CA2065043C - Method for prioritizing selectively discarding and multiplexing differing traffic type fast packets - Google Patents
Method for prioritizing selectively discarding and multiplexing differing traffic type fast packetsInfo
- Publication number
- CA2065043C CA2065043C CA002065043A CA2065043A CA2065043C CA 2065043 C CA2065043 C CA 2065043C CA 002065043 A CA002065043 A CA 002065043A CA 2065043 A CA2065043 A CA 2065043A CA 2065043 C CA2065043 C CA 2065043C
- Authority
- CA
- Canada
- Prior art keywords
- fast packets
- packets
- fast
- queue
- packet
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Lifetime
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04J—MULTIPLEX COMMUNICATION
- H04J3/00—Time-division multiplex systems
- H04J3/24—Time-division multiplex systems in which the allocation is indicated by an address the different channels being transmitted sequentially
- H04J3/247—ATM or packet multiplexing
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04Q—SELECTING
- H04Q11/00—Selecting arrangements for multiplex systems
- H04Q11/04—Selecting arrangements for multiplex systems for time-division multiplexing
- H04Q11/0428—Integrated services digital network, i.e. systems for transmission of different types of digitised signals, e.g. speech, data, telecentral, television signals
- H04Q11/0478—Provisions for broadband connections
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/54—Store-and-forward switching systems
- H04L12/56—Packet switching systems
- H04L12/5601—Transfer mode dependent, e.g. ATM
- H04L2012/5629—Admission control
- H04L2012/5631—Resource management and allocation
- H04L2012/5632—Bandwidth allocation
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/54—Store-and-forward switching systems
- H04L12/56—Packet switching systems
- H04L12/5601—Transfer mode dependent, e.g. ATM
- H04L2012/5638—Services, e.g. multimedia, GOS, QOS
- H04L2012/5646—Cell characteristics, e.g. loss, delay, jitter, sequence integrity
- H04L2012/5647—Cell loss
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/54—Store-and-forward switching systems
- H04L12/56—Packet switching systems
- H04L12/5601—Transfer mode dependent, e.g. ATM
- H04L2012/5638—Services, e.g. multimedia, GOS, QOS
- H04L2012/5646—Cell characteristics, e.g. loss, delay, jitter, sequence integrity
- H04L2012/5651—Priority, marking, classes
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/54—Store-and-forward switching systems
- H04L12/56—Packet switching systems
- H04L12/5601—Transfer mode dependent, e.g. ATM
- H04L2012/5672—Multiplexing, e.g. coding, scrambling
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/54—Store-and-forward switching systems
- H04L12/56—Packet switching systems
- H04L12/5601—Transfer mode dependent, e.g. ATM
- H04L2012/5678—Traffic aspects, e.g. arbitration, load balancing, smoothing, buffer management
- H04L2012/5679—Arbitration or scheduling
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y10—TECHNICAL SUBJECTS COVERED BY FORMER USPC
- Y10S—TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y10S370/00—Multiplex communications
- Y10S370/901—Wide area network
- Y10S370/902—Packet switching
Abstract
A queueing and dequeueing mechanism for use in an integrated fast packet network, wherein fast packets from differing traffic types are multiplexed with one another through use of a weighted round-robin bandwidth allocation mechanism (517).
Fast packets within a particular traffic type are selected for transmission through use of a head of line priority service (514), a packet discard mechanism (516), or both. The weighted round-robin bandwith allocation mechanism functions, in part, based up-on a credit counter for each queue group that represents a particular traffic type.
Fast packets within a particular traffic type are selected for transmission through use of a head of line priority service (514), a packet discard mechanism (516), or both. The weighted round-robin bandwith allocation mechanism functions, in part, based up-on a credit counter for each queue group that represents a particular traffic type.
Description
WO 92/01345 PCI~/US91/044~4 20`B50~3 ~ METHOD FOR PRIORITIZING, SELECTIVELY DISCARDING, AND MULTIPLEXING DIFFERING TRAFFIC TYPE FAST
PACKETS
Tech n ical Field This invention relates generally to integrated fast packet networks, and more particularly to prioritization, at a network trunk node, of fast packets of a common traffic type, selective discarding of certain fast 15 packets, and bandwidth allocation through multiplexing ~~ =
of fast packets from differing traffic type groups.
~ack~round of the Invention Typical integrated fast packet networks carry at least three classes of traffic: continuous bit-stream oriented (CBO), voice, and data. Fig. 1A-C illustrates the bandwidth characteristics and requirements of these differing traffic types, and their attributes are summarized as follows.
CBO: Packets from individual sources are fairly well-behaved and arrive at the internodal trunk queues more or less periodically. The peak rate (Rpeak) of multiplexed CBO sources is the same as the average rate (Ravg)~ and the trunk bandwidth required (Rreqd) is somewhat larger to keep the queueing delays small.
Since Rpeak ' Rreqd, no statistical gain is obtained in this case. CBO streams are sensitive to severe fluctuations in queueing delay and to packet losses since 3~ both of these cause a loss in synchronization at the WO 92/01345 PCI~/US91/04444 6SQ4~ 2 ~
, .~
receiver. Packets from CBO sources with large packetization times have large delays when multiplexed together as opposed to packets from sources with small packetization times. When multiplexed together, the 5 delays are dominated by the large packetization time sources .
Packet Voice (with speech activity detection): The rate of multiplexed voice depends on the number of sources simultaneously in talk spurt and fluctuates 10 between the maximum rate (Rpeak) and zero. The average rate (Ravg) is less than half of Rpeak (for conversational speech). The required rate (Rreqd) can be made to lie in between these two rates (rather than being made equal to Rpeak). making statistical gain (i.e., 15 Rpeak/Rreqd) possible. Rreqd is chosen to keep the maximum delay and the packet loss rate under the given limits (a small loss is acceptable since degradation in voice quality can remain within acceptable limits).
Packets with excessive delays (typically a few 20 hundred millisecs) are also dropped at the destination based on an end-to-end delay limit, since voice is delay sensitive. This type of dropping results in (with a high probability) several packets being consecutively lost from the same voice call and degrades the fidelity of the 2~ received voice signal seriously.
Framed Data: This type of traffic can have large fluctuations in the rate and in the difference between Rpeak and Ravg Rreqd is chosen to keep the end-to-end average frame delays acceptably small. Since the loss 30 of a single fast packet results in of an entire frame, it is not desirable to drop packets.
However, since data traffic is bursty in nature, the loss of packets due to the non-availability of buffers cannot be prevented. Additionally, data traffic from different 35 sources have differing quality of service (QOS) WO 92/01345 PCr/US91/04444 ~ {~65043 requirements; e.g. interactive traffic, is relatively delay sensitive, whereas file transfers are less so.
Integrated networks considered hera rely on fast packet technology to transport such packets between end systems. Before the flow of packets between the end systems begins, a connection (or virtual circuit) is established between them. This connection determines the path (i.e., the nodes and internodal trunks) that the fast packets will follow from end to end. Figure 2 depicts a switch typically used at an intermediate node, that receives fast packets from one or more input trunks and switches them to one or more output trunks.
Packets coming in on an internodal trunk from each connection have a unique field in the header, called the Logical Channel Number (LCN), that corresponds to a logical channel on that trunk (see Fig. 3A). At the time of the establishment of the connection, a table is updated (at each node) that contains entries for the output Port number, the Queue ID (QID) of the output queue, and the new LCN. The LCN of each incoming packet is examined and this address is translated (by access to the table) to the output Port number, QID (for an output queue), and the new LCN for the next internodal trunk (see Fig. 3B).
The packets from the various output queues are transmitted (or discarded) in an order determined by the particular queueing discipline used in the network.
(Though this invention is described with this environment in mind, the queueing discipline structure and methods described herein are applicable to a broad class of connection oriented or connection-less networks and are not limited to an implementation using LCNs and table lookups).
For example, one technique for queueing packets for transmission on an internodal trunk uses a first-in-` ;2065~3 4 first-out (FIFO) queue. Multiplexed voice and data traffic, however, experience overload periods when the instantaneous rate of packets entering a trunk becomes larger than Rreqd (which corresponds to the trunk 5 bandwidth allocated to that traffic). The overload periods for voice can be expected to last from 10 ms to 1000 ms. For data, the periods can last from a few ms to several seconds. CBO sources are sensitive to fluctuations in queueing delays that var,v between 1 ms 10 and 32 ms per hop in typical T1 networks (when CBO
sources alone are multiplexed into the same queue).
Clearly, long overload periods for voice and data caused by a simple FIFO queue could make CBO queueing delay fluctuations as long as the overload periods, and would 15 cause the end-to-end delays to be unacceptable high, since a longer smoothing delay would be required at the destination to prevent gaps from occurring during the bit-stream. Similarly, when data goes into overload, it dominates the trunk bandwidth and causes voice quality 20 to deteriorate (since it would cause voice packets to be delayed or dropped). These clearly undesirable characteristics occur when all the traffic enters one queue.
Another approach uses pure Head Of Line Priority 25 (HOLP~ to give data priority over voice. However, HOLP
does not solve the problem of data and voice queues aKecting the QOS of each other and of CBO when they go into overload. In addition, CBO, voice, and data queues have typical busy periods (1 ms to 32 ms for CBO, 10 ms 30 to 1 sec for voice, and up to a few secs for data) that are not only diKerent but also large enough to cause one type of traffic to dominate the link bandwidth for relatively large amounts of time. For example, if CBO is given highest priority, even though it never goes into 35 overload, a long busy period (say 32 ms) could affect the WO 92/01345 PCI/US91~04444 ! , _ ",. ~ ; 2 0 6 5 0 4 quality of the voice connections because voice packets could be blocked for 32 ms at that hop. If HOLP was used and voice was given either the highest or second highest priority ~it cannot be given the lowest as it is delay sensitive), the voice queue would just take up more bandwidth when it went into overload. Instead of the voice queue growing and losing packets (via a discard mechanism), it would (unfairly) affect the quality of service of the lower priority queues.
Movable boundary schemes for multiplexing two traffic types, voice and data, have also been proposed. In such schemes, the multiplex structure consists of frames that are made up of transmission time slots. At the start of the frame, there are S time slots reserved for voice and the next N slots for data. At the start of the frame, if there are S or more voice packets awaiting transmission, they are loaded into the S time slots and data packets are loaded into the remaining N slots (if there are more than N they are queued). If there are less than S voice packets at the start of the frame, data packets are permitted to use the extra timeslots (thus leading to the term "movable boundary"). Such methods neither consider priorities within a class of traffic nor do they propose an efficient discard mechanism.
In another scheme, timers are used to implement the movable boundary between voice and data queues and a block dropping scheme that assumes embedded coding is used for voice. One approach in particular proposes four classes of traffic (with HOLP within each class if required), alonQ with a discard mechanism for the class corresponding to voice traffic. The movable boundary scheme is implemented by limiting the number of packets transmitted in succession from any traffic class.
06SO~1~ 6 Queueing structures for virtual circuit (VC) based networks have also been proposed. The queueing disciplines described in these methods separates the VCs into priority classes and uses round-robin 5 scheduling between VC's. This discipline is more appropriate for conventional networks that use link-by-link fiow control (as versus fast packet networks that use end-to-end flow control), since the flow into a queue can be tightly controlled. In addition, this discipline 10 does not address the problem of integrating differant traffic types to match quality of service requirements and are geared instead to providing fair service to VCs by serving them in a round-robin order before they are multiplexed. This would not work in the environment of 15 the invention since CBO and voice sources cannot be flow controlled and the HOLP discipline between CBO, voice, and data is insufficient (as described earlier).
Packet discarding has also been proposed to relieve congestion in an integrated voice/data network. In such 20 mechanisms, new incoming packets are discarded if the queues becomes excessive (i.e., filled). A disadvantage of this mechanism is that it discards packets without considering the importance of the packet. Using speech coding and speech activity detection, some of ths 25 packets contain particularly important information. To reduce the impact of this sort of discarding, a selective packet discarding technique has been proposed. Only the packets considered to be less important for voice waveform reconstruction are discarded in face of 30 congestion, which is identified by the queue depth or the number of calls in talk spurt (i.e., packet arrival rate).
In the selective packet discarding, packets are classified into two discard priorities. Both schemes described above use only a single threshold for tha 35 indication of congestion.
WO 92/01345 it `~ 'PCI`/US91/04444 ~ 7 ~ ~206S043 Pursuant to another packet discarding mechanism, if an output queue of an internodal trunk is full, an incoming packet of the queue is dropped. If the queue is not full but the queue depth exceeds a given threshold and the incoming packet is marked droppable, the packet is also d~opped. Otherwise, the packet is placed in the queue waiting for transmission. An alt~rnativ~ for dropping marked packets, referred to as "output dropping" has also been suggested, in which all incoming 10 packets may be placed in the queue if the queue is not full. When a marked packet moves up to the head of the queue and is ready for transmission, the threshold is checked, and the packet will be dropped or transmitted accordingly.
None of the above described methodologies provide fully satisfactory performance, particularly when applied in an integrated fast packet network that supports transmission of varying traffic types.
20 Summary of the Invention These needs and others are substantially met through provision of the fast packet priority queueing, selective discarding, and bandwidth allocation 25 methodology disclosed herein. Pursuant to one embodiment in accordance with the invention, fast pack~ts of differing traffic types are prioritized pursuant to differing prioritization methods vis-a-vis one another. The prioritized packets from each group are 30 then multiplexed and transmitted. For example, a first prioritization method can utilize HOLP, and a second prioritization method can rely upon packet discarding. In the alternative, these approaches can be combined in various ways; for example, a first group of fast packets 3~ for a particular traffic type can be prioritized using 1~-2065o~3 8 HOLP alone, and a second group for differing traffic type can be prioritized using both a packet discard mechanism and a HOLP methodology.
In another embodiment, the multiplexing can be 5 accomplished through USB of a weighted round-robin bandwidth allocation protocol.
Viewed another way, pursuant to one embodiment of the invention, fast packets for differing traffic types are received and prioritized as a function, at least in 10 part, of a first prioritization method, and fast packets that are queued pursuant to the first prioritization method are then multiplexed in accordance with a bandwidth allocation protocol, such as weighted round-robin. In one embodiment, the first prioritization 15 method could be, for example, HOLP.
In yet another embodiment in accordance with the invention, fast packets for differing traffic types are multiplexed by queuing fast packets for each traffic type in a separate queue, such that queues that contain fast 20 packets for a particular traffic type are grouped together discrete from other traffic types. A first group of queues can then be selected, and a credit count associated with the first group of queues can be examined to determine whether any credit remains 25 therein. If so, one fast packet buffered within a queue corresponding to the highest priority queue in the group that has a packet can be removed and transmitted. Upon transmission, the credit count can be altered in a manner corresponding to the length of the fast packet. For 30 sxample, in one embodiment, the credit count can be decremented by an amount equal to the number of bytes in the transmitted fast packet. The bandwidth allocation is achieved by incrementing the credit count for a queue group by different amounts for different groups.
3~
PACKETS
Tech n ical Field This invention relates generally to integrated fast packet networks, and more particularly to prioritization, at a network trunk node, of fast packets of a common traffic type, selective discarding of certain fast 15 packets, and bandwidth allocation through multiplexing ~~ =
of fast packets from differing traffic type groups.
~ack~round of the Invention Typical integrated fast packet networks carry at least three classes of traffic: continuous bit-stream oriented (CBO), voice, and data. Fig. 1A-C illustrates the bandwidth characteristics and requirements of these differing traffic types, and their attributes are summarized as follows.
CBO: Packets from individual sources are fairly well-behaved and arrive at the internodal trunk queues more or less periodically. The peak rate (Rpeak) of multiplexed CBO sources is the same as the average rate (Ravg)~ and the trunk bandwidth required (Rreqd) is somewhat larger to keep the queueing delays small.
Since Rpeak ' Rreqd, no statistical gain is obtained in this case. CBO streams are sensitive to severe fluctuations in queueing delay and to packet losses since 3~ both of these cause a loss in synchronization at the WO 92/01345 PCI~/US91/04444 6SQ4~ 2 ~
, .~
receiver. Packets from CBO sources with large packetization times have large delays when multiplexed together as opposed to packets from sources with small packetization times. When multiplexed together, the 5 delays are dominated by the large packetization time sources .
Packet Voice (with speech activity detection): The rate of multiplexed voice depends on the number of sources simultaneously in talk spurt and fluctuates 10 between the maximum rate (Rpeak) and zero. The average rate (Ravg) is less than half of Rpeak (for conversational speech). The required rate (Rreqd) can be made to lie in between these two rates (rather than being made equal to Rpeak). making statistical gain (i.e., 15 Rpeak/Rreqd) possible. Rreqd is chosen to keep the maximum delay and the packet loss rate under the given limits (a small loss is acceptable since degradation in voice quality can remain within acceptable limits).
Packets with excessive delays (typically a few 20 hundred millisecs) are also dropped at the destination based on an end-to-end delay limit, since voice is delay sensitive. This type of dropping results in (with a high probability) several packets being consecutively lost from the same voice call and degrades the fidelity of the 2~ received voice signal seriously.
Framed Data: This type of traffic can have large fluctuations in the rate and in the difference between Rpeak and Ravg Rreqd is chosen to keep the end-to-end average frame delays acceptably small. Since the loss 30 of a single fast packet results in of an entire frame, it is not desirable to drop packets.
However, since data traffic is bursty in nature, the loss of packets due to the non-availability of buffers cannot be prevented. Additionally, data traffic from different 35 sources have differing quality of service (QOS) WO 92/01345 PCr/US91/04444 ~ {~65043 requirements; e.g. interactive traffic, is relatively delay sensitive, whereas file transfers are less so.
Integrated networks considered hera rely on fast packet technology to transport such packets between end systems. Before the flow of packets between the end systems begins, a connection (or virtual circuit) is established between them. This connection determines the path (i.e., the nodes and internodal trunks) that the fast packets will follow from end to end. Figure 2 depicts a switch typically used at an intermediate node, that receives fast packets from one or more input trunks and switches them to one or more output trunks.
Packets coming in on an internodal trunk from each connection have a unique field in the header, called the Logical Channel Number (LCN), that corresponds to a logical channel on that trunk (see Fig. 3A). At the time of the establishment of the connection, a table is updated (at each node) that contains entries for the output Port number, the Queue ID (QID) of the output queue, and the new LCN. The LCN of each incoming packet is examined and this address is translated (by access to the table) to the output Port number, QID (for an output queue), and the new LCN for the next internodal trunk (see Fig. 3B).
The packets from the various output queues are transmitted (or discarded) in an order determined by the particular queueing discipline used in the network.
(Though this invention is described with this environment in mind, the queueing discipline structure and methods described herein are applicable to a broad class of connection oriented or connection-less networks and are not limited to an implementation using LCNs and table lookups).
For example, one technique for queueing packets for transmission on an internodal trunk uses a first-in-` ;2065~3 4 first-out (FIFO) queue. Multiplexed voice and data traffic, however, experience overload periods when the instantaneous rate of packets entering a trunk becomes larger than Rreqd (which corresponds to the trunk 5 bandwidth allocated to that traffic). The overload periods for voice can be expected to last from 10 ms to 1000 ms. For data, the periods can last from a few ms to several seconds. CBO sources are sensitive to fluctuations in queueing delays that var,v between 1 ms 10 and 32 ms per hop in typical T1 networks (when CBO
sources alone are multiplexed into the same queue).
Clearly, long overload periods for voice and data caused by a simple FIFO queue could make CBO queueing delay fluctuations as long as the overload periods, and would 15 cause the end-to-end delays to be unacceptable high, since a longer smoothing delay would be required at the destination to prevent gaps from occurring during the bit-stream. Similarly, when data goes into overload, it dominates the trunk bandwidth and causes voice quality 20 to deteriorate (since it would cause voice packets to be delayed or dropped). These clearly undesirable characteristics occur when all the traffic enters one queue.
Another approach uses pure Head Of Line Priority 25 (HOLP~ to give data priority over voice. However, HOLP
does not solve the problem of data and voice queues aKecting the QOS of each other and of CBO when they go into overload. In addition, CBO, voice, and data queues have typical busy periods (1 ms to 32 ms for CBO, 10 ms 30 to 1 sec for voice, and up to a few secs for data) that are not only diKerent but also large enough to cause one type of traffic to dominate the link bandwidth for relatively large amounts of time. For example, if CBO is given highest priority, even though it never goes into 35 overload, a long busy period (say 32 ms) could affect the WO 92/01345 PCI/US91~04444 ! , _ ",. ~ ; 2 0 6 5 0 4 quality of the voice connections because voice packets could be blocked for 32 ms at that hop. If HOLP was used and voice was given either the highest or second highest priority ~it cannot be given the lowest as it is delay sensitive), the voice queue would just take up more bandwidth when it went into overload. Instead of the voice queue growing and losing packets (via a discard mechanism), it would (unfairly) affect the quality of service of the lower priority queues.
Movable boundary schemes for multiplexing two traffic types, voice and data, have also been proposed. In such schemes, the multiplex structure consists of frames that are made up of transmission time slots. At the start of the frame, there are S time slots reserved for voice and the next N slots for data. At the start of the frame, if there are S or more voice packets awaiting transmission, they are loaded into the S time slots and data packets are loaded into the remaining N slots (if there are more than N they are queued). If there are less than S voice packets at the start of the frame, data packets are permitted to use the extra timeslots (thus leading to the term "movable boundary"). Such methods neither consider priorities within a class of traffic nor do they propose an efficient discard mechanism.
In another scheme, timers are used to implement the movable boundary between voice and data queues and a block dropping scheme that assumes embedded coding is used for voice. One approach in particular proposes four classes of traffic (with HOLP within each class if required), alonQ with a discard mechanism for the class corresponding to voice traffic. The movable boundary scheme is implemented by limiting the number of packets transmitted in succession from any traffic class.
06SO~1~ 6 Queueing structures for virtual circuit (VC) based networks have also been proposed. The queueing disciplines described in these methods separates the VCs into priority classes and uses round-robin 5 scheduling between VC's. This discipline is more appropriate for conventional networks that use link-by-link fiow control (as versus fast packet networks that use end-to-end flow control), since the flow into a queue can be tightly controlled. In addition, this discipline 10 does not address the problem of integrating differant traffic types to match quality of service requirements and are geared instead to providing fair service to VCs by serving them in a round-robin order before they are multiplexed. This would not work in the environment of 15 the invention since CBO and voice sources cannot be flow controlled and the HOLP discipline between CBO, voice, and data is insufficient (as described earlier).
Packet discarding has also been proposed to relieve congestion in an integrated voice/data network. In such 20 mechanisms, new incoming packets are discarded if the queues becomes excessive (i.e., filled). A disadvantage of this mechanism is that it discards packets without considering the importance of the packet. Using speech coding and speech activity detection, some of ths 25 packets contain particularly important information. To reduce the impact of this sort of discarding, a selective packet discarding technique has been proposed. Only the packets considered to be less important for voice waveform reconstruction are discarded in face of 30 congestion, which is identified by the queue depth or the number of calls in talk spurt (i.e., packet arrival rate).
In the selective packet discarding, packets are classified into two discard priorities. Both schemes described above use only a single threshold for tha 35 indication of congestion.
WO 92/01345 it `~ 'PCI`/US91/04444 ~ 7 ~ ~206S043 Pursuant to another packet discarding mechanism, if an output queue of an internodal trunk is full, an incoming packet of the queue is dropped. If the queue is not full but the queue depth exceeds a given threshold and the incoming packet is marked droppable, the packet is also d~opped. Otherwise, the packet is placed in the queue waiting for transmission. An alt~rnativ~ for dropping marked packets, referred to as "output dropping" has also been suggested, in which all incoming 10 packets may be placed in the queue if the queue is not full. When a marked packet moves up to the head of the queue and is ready for transmission, the threshold is checked, and the packet will be dropped or transmitted accordingly.
None of the above described methodologies provide fully satisfactory performance, particularly when applied in an integrated fast packet network that supports transmission of varying traffic types.
20 Summary of the Invention These needs and others are substantially met through provision of the fast packet priority queueing, selective discarding, and bandwidth allocation 25 methodology disclosed herein. Pursuant to one embodiment in accordance with the invention, fast pack~ts of differing traffic types are prioritized pursuant to differing prioritization methods vis-a-vis one another. The prioritized packets from each group are 30 then multiplexed and transmitted. For example, a first prioritization method can utilize HOLP, and a second prioritization method can rely upon packet discarding. In the alternative, these approaches can be combined in various ways; for example, a first group of fast packets 3~ for a particular traffic type can be prioritized using 1~-2065o~3 8 HOLP alone, and a second group for differing traffic type can be prioritized using both a packet discard mechanism and a HOLP methodology.
In another embodiment, the multiplexing can be 5 accomplished through USB of a weighted round-robin bandwidth allocation protocol.
Viewed another way, pursuant to one embodiment of the invention, fast packets for differing traffic types are received and prioritized as a function, at least in 10 part, of a first prioritization method, and fast packets that are queued pursuant to the first prioritization method are then multiplexed in accordance with a bandwidth allocation protocol, such as weighted round-robin. In one embodiment, the first prioritization 15 method could be, for example, HOLP.
In yet another embodiment in accordance with the invention, fast packets for differing traffic types are multiplexed by queuing fast packets for each traffic type in a separate queue, such that queues that contain fast 20 packets for a particular traffic type are grouped together discrete from other traffic types. A first group of queues can then be selected, and a credit count associated with the first group of queues can be examined to determine whether any credit remains 25 therein. If so, one fast packet buffered within a queue corresponding to the highest priority queue in the group that has a packet can be removed and transmitted. Upon transmission, the credit count can be altered in a manner corresponding to the length of the fast packet. For 30 sxample, in one embodiment, the credit count can be decremented by an amount equal to the number of bytes in the transmitted fast packet. The bandwidth allocation is achieved by incrementing the credit count for a queue group by different amounts for different groups.
3~
4~ PCI/US91104444 ~ 9 ~ 0 6 ~ O ~ 3 Rrief Description of the Drawin~s Figs. 1A-C comprises a depiction of prior art fast packet traffic types;
Fig. 2 comprises a block diagram depiction of a prior art fast packet switch with input analysis and output queue;
Figs. 3A-B comprises a prior art depiction of a information header as initially received and as subsequently processed by the analysis block for the prio r art switch;
Fig. 4 comprises a depiction of a queuing and multiplexing process in accordance with the invention;
Fig. 5 comprises a block diagram depiction of one possible queueing and multiplexing process in accordance with the invention;
Fig. 6 comprises a block diagram depiction of an alternative embodiment of a queuing and multiplexing arrangement in accordance with the invention;
Figs. 7-9 comprise flow diagrams depicting operation of queuing, discarding, and multiplexing aspects of an internodal switch in accordance with the invention.
2~ r)etailed Description of the Preferred Embodiment Fig. 4 provides a general depiction of the enqueueing and dequeueing process contemplated herein (400). Following an appropriate enqueueing process (401), fast packets from various traffic types, such as CBO, voice, and framed data, are buffered in appropriate queues, which queues are segregated into groups (402) that correspond to the differing traffic types~ In this example, the first group (403) buffers CBO fast packets, the second group (404) buffers voice fast packets, and WO 92iO134s PCr/US91/04444 ~065043 the third group (405) buffers framed data fast packets.
(Additional description of the various queues will be provided below where appropriate.) The buffered fast packets are then dequeued through use of an appropriate dequeueing process (406) and provided to an internodal trunk to support continued transmission of the dequeued fast packets.
Turning now to Fig. 5, fast packets of different traffic types are routed by a switch and enqueueing process (505) (using the header information) to the appropriate queue. In this embodiment, voice packets are queued into a voice queue (506). The remaining two traffic type groups (CBO and framed data) are each provided with multiple queues (507-9 and 511-13, 1 5 respectively).
These multiple queues provide the differing quality of service that is required by CBO and data traffic with different characteristics. CBO sources typically have varying packetization times and packet lengths; hence, this syst~m will separate out packets from different sources into more than one queue because the worst case delay for any queue is determined by the largest packetization time for sources multiplexed into that queue. For data, the sources can generate short length data frames (for example, from an interactive terminal application) or large amounts of information (such as file transfers). Both types are not placed in a single queùe since an interactive user expects quick response times whereas a file transfer can take longer.
A HOLP service discipline (514) between different data qùeues (with high priority (511) to short frames) or CBO queues (with a higher priority for queues (507) for sources with low burst size (packetization times)) provides an appropriate quality of service for the different types of traffic in those queues. For example, 11 206~043 t.,. ~ ;--with three CBO queues (507-9), all packets from sources with packetization times less than 5 ms could be put into the highest priority queue (507), sources with packetization time between 5 ms and 25 ms in the intermediate priority queue (508) and sources with packetization times larger than 25 ms in the lowest priority queue (509). For data, the allocation is done similarly based on average frame lengths. As an example, sources with frame sizes less than 64 bytes could be assigned to the highest priority queue (511), frame sizes between 64 and 512 bytes to the intermediate priority queue (512), and larger than 512 bytes to the lowest priority queue (513).
In general, a HOLP discipline tends to give superior service to the high priority queues (507 and 511) at the expense of degrading the performance of ths lower priority queues. In this embodiment, however, the lower priority queues do not experience a severe degradation because the higher priority CBO queues and short data frames in the higher priority data queues have short busy periods.
The voice queue has traffic which is similar (even with different coding rates) and therefore only a single queue is needed.
It will be noted that, in this embodiment, a head of line priority mechanism has not been provided for the voice fast packets, since only one voice fast packet queue (506) has been provided. Instead, a packet discarder (516) has been provided. The functioning of this packet discarder will now be described in more detail with reference to Fig. 6.
Fig. 6 depicts an altered embodiment wherein the voice and framed data fast packets are processed through a packet discard mechanism as well as a HOLP
service. There are two bits in the header of each fast ;~ 2~5~3 12 ~
packet that indicates a discard priority. The data packets are ciassified into four discard priorities: first, second, third, and last discarded. The voice packets are classified into three discard priorities: first, second, 5 and last discarded. The discard priority is assigned to a packet based on its importance in reconstructing the voice waveform or its effect on the number of retransmitted data frames. A number of water-marks (A-C) corresponding to the discard priorities are 10 provided in the voice and data queues. With this discarding mechanism, all arriving packets are allowed to get into the queue. If the queue is filled (i.e., the last water-mark is exceeded), the packet in the front of the queue is dropped no matter what the discard priority of 15 this packet is. If the queue is not full, packets can be discarded from the front of the queue (based on the water-marks, queue depth, and discard priority of the packet at the front of the queue) at the instant of packet departure from the queue or at the instant of packet 20 arrival to the queue. These are termed departure time discarding and arrival time discarding.
In departure time discarding, the packet discarding is done prior to the departure of a packet. Once the trunk is ready for a packet transmission, the packet in the 25 front of the queue is checked. The discard priority of this packet is read and the current queue depth is compared with the water-mark related to the discard priority of this packet. If the queue depth exceeds the water-mark, the packet is discarded. The next packet in 30 the front of queue is then examined. The procedure repeats until the queue depth does not exceed the water-mark related to the discard priority of a packet in the front of the queue. Finally, this packet is transmitted.
In arrival time discarding, the packet discarding is 35 done after the arrival of a new packet. After the arrival WO 92/0134~ PCI'/US91/04444 13 206~0~3 of a new packet, the discard priority of the packet in the front of the queue is read and the current queue depth is compared with the water-mark related to the discard priority of that packet. If the queue depth exceeds the 5 water-mark, the packet is discarded. Unlike departure time discarding, no more packets are checked for discarding. Once the trunk is ready for a packet transmission, the packet in the front of the queue is transmitted. (Packets in a higher priority queu~ can be 10 discarded, not only based on the queue depth of the queue they are queued in, but also based on the depth of some other queue in the queuing structure. The measursment of the queue depth can be an instantaneous queue depth, or an average queue depth (over some window of time).) Finally, returning to Fig. 5, this embodiment provides a weighted round-robin (WRR) packet selector (517) to provide a bandwidth allocation service. In effect, the packet selector functions to multiplex fast packets from the various queues and provide them to the 20 trunk for continued transmission.
With WRR, the server serves the queues cyclically in a prescribed order. It examines each queue a specified number of times (proportional to its weight). If all the queues have packets to send, WRR has the effect of 25 dividing the trunk bandwidth among the queues in ths ratio of the weights. When some queues are idle, WRR
does not waste bandwidth but continues to examine the queues in the prescribed order until it finds a packet.
WRR is similar to a conventional TDM system except that 30 timeslots do not go idle - allocated bandwidth unused by one queue can be used by another queue.
To efficiently multiplex additional arbitrary traffic types, one would add a new queue group (that will be served by WRR) if it is found that the traffic type 3~ cannot be placed into an existing queue group due to 206~0~3 14 bandwidth usage incompatibilities with the other traffic in that group. This can be done by appropriately characterizing the rate fluctuations and the busy period behavior of the traffic type when identical sources of 5 that type are multiplexed together. Following this, HOLP
queues should be created within that group to finely adjust the quality of service (i.e., delay and packet loss) of sources of the same type but with different parameters (note that if these parameters are 10 significantly different for some sources, new traffic types should be defined which are then placed in different queue groups).
The methodology described below determines the order of service of the different queues so that the 15 overall service discipline is a hybrid of WRR and HOLP.
Through use of this approach, the queue groups can have variable length packets and still be allocated the required fraction of trunk bandwidth on the average.
Also, given that the smallest unit of transmission is a 20 packet, the method attempts to interleave packets as finely as possible so that delay jitter for different queue groups is reduced.
As explained below, a credit counter is maintained for each queue group and this counter is incremented by 25 a certain number of credits at each scan. A queue group can transmit packets when scanned if the number of credits it has is greater than zero. In this embodiment, the credits are measured in bytes. Every time a packet is transmitted, the credit counter is decremented by the 30 number of bytes in the packet. The credit counter is allowed to go negative (if a large size packet is transmitted) but is never allowed to exceed a (positive) maximum value.
The method is described in terms of the enqueueing 35 and dequeueing processes referred to above that perform WO 92/01345 PCI/US91/04~44 1~ 2065043 the task of queueing packets into the output queues and removing a packet from an output queue, respectively. In this description, there is one enqueueing process that moves packets from a switch output port to output queues and one dequeueing process that moves packets from output queues to the output trunk.
A description of the variables used in the following explanation is as follows:
LCN: Logical Channel Number that is in the packet header at an input.
QID: The address of an output queue corresponding to the LCN, obtained from a lookup table DP: The discard priority of a packet.
QUEUE LEN(QID): The number of packets queued at some instant in a queue with the address of QID.
WATERMARK(QID,DP): If the QUEUE LEN(QID) for some queue with address QID is greater than WATERMARK(QID,DP), then a packet with discard priority DP can be discarded.
CF~EDIT[J]: The valu~ of the credit counter for queue group J.
CMAXEJ]: The maximum value that CREDIT~J] can attain. This is also the number of credits given to queue group J at each scan. CREDIT[J] is set to CMAX[J]
initially.
MAX-Q-GRPS: The total number of queue groups that the enqueueing and dequeueing processes serve.
The proper selection of CMAX[J] allows the groups to have variable packet lengths and also enables the system to interleave traffic from the queue groups in a fine grained manner.
These are precomputed as follows. We define:
L[J] = average packet length (in bytes) in queue group J
5 0 ~ 3 16 f[J] 5 fraction of trunk bandwidth to be allocated to group J when all groups have packets CT. CMAX[1] + CMAX[2] + . . . + CMAX[MAX-Q-GRPS].
Since f[J] = CMAXD]/CT, we have the relation 5 CMAX[I]/CMAX[J] ~ f[l]/f[J] for all l,J. To obtain the maximum interleaving, the maximum number of packets that can be transmitted together (when other groups have credits and packets waiting) is made as small as possible (i.e., 1 packet). The maximum number of 10 packets that can be transmitted together is max(CMAX[J]/L[J]). CMAX~J]/L[J] = f[J]~CT/L~J] attains its maximum when f[J]/L[J] attains its maximum. The method described below makes use of these relationships to compute the appropriate CMAX[J]s.
1 = J for which f[J]/L[J] is maximum;
CMAX[I] = L[l];
for J - 1 to MAX-Q-GRPS do CMAX[J] = f[J]*CMAX[I]/f[l].
The enqueueing process (700) appears in Fig. 7. QID
20 is first set as the QID identified in the packet field (701). If arrival time discarding is being used (702), the discard priority (DP) of the packet in the front of the queue is determined from the corresponding packet field (703). The appropriate queue length is then compared to 25 the appropriate water-mark (704), and if the queue length exceeds the relevant watermark, the packet in the front of the queue is discarded (706). Otherwise, the incoming packet is enqueued (708) and the queue length variable is increased by one packet (707). If arrival 30 time discarding is not being used (702), then the process simply enqueues the incoming packet (708) and again increases the queue length by 1 (707).
Referring now to Fig. 8, the dequeueing process appears generally as depicted by reference numeral 800.
3~ To begin, CREDIT [J] is established as zero for all J's WO 92/01345 PCI~/US91/04444 ~ 17 `~1 7_ ~6~043 (801). Next, variable J is initialized (802~. The credit counter for the selected queue group is then established (803), and the prbcess determines whether positive credit exists for the queue group in question and whether there are any fast packets queued for transmission in the relevant group (804). If true, the process scans the next queue in the selected group (th~ particular queue being selected in order of priority) (805). When a packet is found (806) the process continues as described in Fig.
9 (807).
QID is established as the ID of the queue in which the packet has been found (901). The appropriate packet is then obtained (902) and the relevant queue length is decremented by one (903) to reflect removal of ths selected packet. If departure time discarding is being used (904), the discard priority of the packet is obtained (906) and a determination made as to whether the queue length plus one exceeds the relevant watermark (907).
If it does, the first packet in the queue is discarded (908) and a determination then made as to whether the queue length exceeds zero (909). If it does, the next packet is fetched (902) and the process continues as described above. Otherwise, if the queue length does not exceed zero, the process returns.
Assuming the queue length plus one does not exceed the relevant watermark (907), the selected packet will be transmitted (911) and the credit count will be decremented by an amount equal to the number of bytes in the transmitted packet (912). the queue length and the credit count will then be compared to a threshold - (such as "0")(913), and if appropriate amounts remain for both, a new packet will be fetched (902). Otherwise, the process will return .
Returning to Fig. 8, the process returns to determine whether any credit remains for this particular WO 92/01345 PCI'/US91/04444 ~56~043 18 queue group (804). Eventually, one of two events will occur. Either no packets will be found (806) and all of the queues in the group will have been scanned (811), or the credit for the group (804) will be completely 5 cieared. In either of these two events, the process will determine whether all queue groups have yet been scanned (812~. If not, the queue group identifier will be incremented (813) and the above described process will repeat for the next queue group. Otherwise, when all of 10 the queue groups have been examined, a determination will be made as to whether the credit counters for all queues have all become negative (814). If not, the process will begin again with the first queue group (802). Otherwise, appropriate steps will be taken (816 15 and 817) to reset the credit counters for all of the queue groups to an initial level, following which the above described procedures will again repeat.
The invention has been described for an application of the queueing discipline at the output of a packet 20 switch. The same techniques are equally applicable in the case of packet switches where the segregation and queueing on the basis of the traffic type is done at the input of the switch.
In addition, the methods described here are not 25 restricted to integrated networks that carry traffic from only CBO, voice, and data sources. For example, compressed video and image/facsimile traffic would be treated as separate traffic types if the network was to carry them. The bandwidth requirements of compressed 30 video fluctuate over time; when successive screens are not very different (for example, a slowly moving object) the bit rate is relatively small. On the other hand, rapidly moving objects or scene changes raise the instantaneous bandwidth requirements to higher levels.
35 In addition, packets from both video and image/facsimile WO 92/01345 PCr/l~ Og~4~ 3 sources can selectively be dropped at the cost of little or no degradation in the image quality.
What is claimed is:
Fig. 2 comprises a block diagram depiction of a prior art fast packet switch with input analysis and output queue;
Figs. 3A-B comprises a prior art depiction of a information header as initially received and as subsequently processed by the analysis block for the prio r art switch;
Fig. 4 comprises a depiction of a queuing and multiplexing process in accordance with the invention;
Fig. 5 comprises a block diagram depiction of one possible queueing and multiplexing process in accordance with the invention;
Fig. 6 comprises a block diagram depiction of an alternative embodiment of a queuing and multiplexing arrangement in accordance with the invention;
Figs. 7-9 comprise flow diagrams depicting operation of queuing, discarding, and multiplexing aspects of an internodal switch in accordance with the invention.
2~ r)etailed Description of the Preferred Embodiment Fig. 4 provides a general depiction of the enqueueing and dequeueing process contemplated herein (400). Following an appropriate enqueueing process (401), fast packets from various traffic types, such as CBO, voice, and framed data, are buffered in appropriate queues, which queues are segregated into groups (402) that correspond to the differing traffic types~ In this example, the first group (403) buffers CBO fast packets, the second group (404) buffers voice fast packets, and WO 92iO134s PCr/US91/04444 ~065043 the third group (405) buffers framed data fast packets.
(Additional description of the various queues will be provided below where appropriate.) The buffered fast packets are then dequeued through use of an appropriate dequeueing process (406) and provided to an internodal trunk to support continued transmission of the dequeued fast packets.
Turning now to Fig. 5, fast packets of different traffic types are routed by a switch and enqueueing process (505) (using the header information) to the appropriate queue. In this embodiment, voice packets are queued into a voice queue (506). The remaining two traffic type groups (CBO and framed data) are each provided with multiple queues (507-9 and 511-13, 1 5 respectively).
These multiple queues provide the differing quality of service that is required by CBO and data traffic with different characteristics. CBO sources typically have varying packetization times and packet lengths; hence, this syst~m will separate out packets from different sources into more than one queue because the worst case delay for any queue is determined by the largest packetization time for sources multiplexed into that queue. For data, the sources can generate short length data frames (for example, from an interactive terminal application) or large amounts of information (such as file transfers). Both types are not placed in a single queùe since an interactive user expects quick response times whereas a file transfer can take longer.
A HOLP service discipline (514) between different data qùeues (with high priority (511) to short frames) or CBO queues (with a higher priority for queues (507) for sources with low burst size (packetization times)) provides an appropriate quality of service for the different types of traffic in those queues. For example, 11 206~043 t.,. ~ ;--with three CBO queues (507-9), all packets from sources with packetization times less than 5 ms could be put into the highest priority queue (507), sources with packetization time between 5 ms and 25 ms in the intermediate priority queue (508) and sources with packetization times larger than 25 ms in the lowest priority queue (509). For data, the allocation is done similarly based on average frame lengths. As an example, sources with frame sizes less than 64 bytes could be assigned to the highest priority queue (511), frame sizes between 64 and 512 bytes to the intermediate priority queue (512), and larger than 512 bytes to the lowest priority queue (513).
In general, a HOLP discipline tends to give superior service to the high priority queues (507 and 511) at the expense of degrading the performance of ths lower priority queues. In this embodiment, however, the lower priority queues do not experience a severe degradation because the higher priority CBO queues and short data frames in the higher priority data queues have short busy periods.
The voice queue has traffic which is similar (even with different coding rates) and therefore only a single queue is needed.
It will be noted that, in this embodiment, a head of line priority mechanism has not been provided for the voice fast packets, since only one voice fast packet queue (506) has been provided. Instead, a packet discarder (516) has been provided. The functioning of this packet discarder will now be described in more detail with reference to Fig. 6.
Fig. 6 depicts an altered embodiment wherein the voice and framed data fast packets are processed through a packet discard mechanism as well as a HOLP
service. There are two bits in the header of each fast ;~ 2~5~3 12 ~
packet that indicates a discard priority. The data packets are ciassified into four discard priorities: first, second, third, and last discarded. The voice packets are classified into three discard priorities: first, second, 5 and last discarded. The discard priority is assigned to a packet based on its importance in reconstructing the voice waveform or its effect on the number of retransmitted data frames. A number of water-marks (A-C) corresponding to the discard priorities are 10 provided in the voice and data queues. With this discarding mechanism, all arriving packets are allowed to get into the queue. If the queue is filled (i.e., the last water-mark is exceeded), the packet in the front of the queue is dropped no matter what the discard priority of 15 this packet is. If the queue is not full, packets can be discarded from the front of the queue (based on the water-marks, queue depth, and discard priority of the packet at the front of the queue) at the instant of packet departure from the queue or at the instant of packet 20 arrival to the queue. These are termed departure time discarding and arrival time discarding.
In departure time discarding, the packet discarding is done prior to the departure of a packet. Once the trunk is ready for a packet transmission, the packet in the 25 front of the queue is checked. The discard priority of this packet is read and the current queue depth is compared with the water-mark related to the discard priority of this packet. If the queue depth exceeds the water-mark, the packet is discarded. The next packet in 30 the front of queue is then examined. The procedure repeats until the queue depth does not exceed the water-mark related to the discard priority of a packet in the front of the queue. Finally, this packet is transmitted.
In arrival time discarding, the packet discarding is 35 done after the arrival of a new packet. After the arrival WO 92/0134~ PCI'/US91/04444 13 206~0~3 of a new packet, the discard priority of the packet in the front of the queue is read and the current queue depth is compared with the water-mark related to the discard priority of that packet. If the queue depth exceeds the 5 water-mark, the packet is discarded. Unlike departure time discarding, no more packets are checked for discarding. Once the trunk is ready for a packet transmission, the packet in the front of the queue is transmitted. (Packets in a higher priority queu~ can be 10 discarded, not only based on the queue depth of the queue they are queued in, but also based on the depth of some other queue in the queuing structure. The measursment of the queue depth can be an instantaneous queue depth, or an average queue depth (over some window of time).) Finally, returning to Fig. 5, this embodiment provides a weighted round-robin (WRR) packet selector (517) to provide a bandwidth allocation service. In effect, the packet selector functions to multiplex fast packets from the various queues and provide them to the 20 trunk for continued transmission.
With WRR, the server serves the queues cyclically in a prescribed order. It examines each queue a specified number of times (proportional to its weight). If all the queues have packets to send, WRR has the effect of 25 dividing the trunk bandwidth among the queues in ths ratio of the weights. When some queues are idle, WRR
does not waste bandwidth but continues to examine the queues in the prescribed order until it finds a packet.
WRR is similar to a conventional TDM system except that 30 timeslots do not go idle - allocated bandwidth unused by one queue can be used by another queue.
To efficiently multiplex additional arbitrary traffic types, one would add a new queue group (that will be served by WRR) if it is found that the traffic type 3~ cannot be placed into an existing queue group due to 206~0~3 14 bandwidth usage incompatibilities with the other traffic in that group. This can be done by appropriately characterizing the rate fluctuations and the busy period behavior of the traffic type when identical sources of 5 that type are multiplexed together. Following this, HOLP
queues should be created within that group to finely adjust the quality of service (i.e., delay and packet loss) of sources of the same type but with different parameters (note that if these parameters are 10 significantly different for some sources, new traffic types should be defined which are then placed in different queue groups).
The methodology described below determines the order of service of the different queues so that the 15 overall service discipline is a hybrid of WRR and HOLP.
Through use of this approach, the queue groups can have variable length packets and still be allocated the required fraction of trunk bandwidth on the average.
Also, given that the smallest unit of transmission is a 20 packet, the method attempts to interleave packets as finely as possible so that delay jitter for different queue groups is reduced.
As explained below, a credit counter is maintained for each queue group and this counter is incremented by 25 a certain number of credits at each scan. A queue group can transmit packets when scanned if the number of credits it has is greater than zero. In this embodiment, the credits are measured in bytes. Every time a packet is transmitted, the credit counter is decremented by the 30 number of bytes in the packet. The credit counter is allowed to go negative (if a large size packet is transmitted) but is never allowed to exceed a (positive) maximum value.
The method is described in terms of the enqueueing 35 and dequeueing processes referred to above that perform WO 92/01345 PCI/US91/04~44 1~ 2065043 the task of queueing packets into the output queues and removing a packet from an output queue, respectively. In this description, there is one enqueueing process that moves packets from a switch output port to output queues and one dequeueing process that moves packets from output queues to the output trunk.
A description of the variables used in the following explanation is as follows:
LCN: Logical Channel Number that is in the packet header at an input.
QID: The address of an output queue corresponding to the LCN, obtained from a lookup table DP: The discard priority of a packet.
QUEUE LEN(QID): The number of packets queued at some instant in a queue with the address of QID.
WATERMARK(QID,DP): If the QUEUE LEN(QID) for some queue with address QID is greater than WATERMARK(QID,DP), then a packet with discard priority DP can be discarded.
CF~EDIT[J]: The valu~ of the credit counter for queue group J.
CMAXEJ]: The maximum value that CREDIT~J] can attain. This is also the number of credits given to queue group J at each scan. CREDIT[J] is set to CMAX[J]
initially.
MAX-Q-GRPS: The total number of queue groups that the enqueueing and dequeueing processes serve.
The proper selection of CMAX[J] allows the groups to have variable packet lengths and also enables the system to interleave traffic from the queue groups in a fine grained manner.
These are precomputed as follows. We define:
L[J] = average packet length (in bytes) in queue group J
5 0 ~ 3 16 f[J] 5 fraction of trunk bandwidth to be allocated to group J when all groups have packets CT. CMAX[1] + CMAX[2] + . . . + CMAX[MAX-Q-GRPS].
Since f[J] = CMAXD]/CT, we have the relation 5 CMAX[I]/CMAX[J] ~ f[l]/f[J] for all l,J. To obtain the maximum interleaving, the maximum number of packets that can be transmitted together (when other groups have credits and packets waiting) is made as small as possible (i.e., 1 packet). The maximum number of 10 packets that can be transmitted together is max(CMAX[J]/L[J]). CMAX~J]/L[J] = f[J]~CT/L~J] attains its maximum when f[J]/L[J] attains its maximum. The method described below makes use of these relationships to compute the appropriate CMAX[J]s.
1 = J for which f[J]/L[J] is maximum;
CMAX[I] = L[l];
for J - 1 to MAX-Q-GRPS do CMAX[J] = f[J]*CMAX[I]/f[l].
The enqueueing process (700) appears in Fig. 7. QID
20 is first set as the QID identified in the packet field (701). If arrival time discarding is being used (702), the discard priority (DP) of the packet in the front of the queue is determined from the corresponding packet field (703). The appropriate queue length is then compared to 25 the appropriate water-mark (704), and if the queue length exceeds the relevant watermark, the packet in the front of the queue is discarded (706). Otherwise, the incoming packet is enqueued (708) and the queue length variable is increased by one packet (707). If arrival 30 time discarding is not being used (702), then the process simply enqueues the incoming packet (708) and again increases the queue length by 1 (707).
Referring now to Fig. 8, the dequeueing process appears generally as depicted by reference numeral 800.
3~ To begin, CREDIT [J] is established as zero for all J's WO 92/01345 PCI~/US91/04444 ~ 17 `~1 7_ ~6~043 (801). Next, variable J is initialized (802~. The credit counter for the selected queue group is then established (803), and the prbcess determines whether positive credit exists for the queue group in question and whether there are any fast packets queued for transmission in the relevant group (804). If true, the process scans the next queue in the selected group (th~ particular queue being selected in order of priority) (805). When a packet is found (806) the process continues as described in Fig.
9 (807).
QID is established as the ID of the queue in which the packet has been found (901). The appropriate packet is then obtained (902) and the relevant queue length is decremented by one (903) to reflect removal of ths selected packet. If departure time discarding is being used (904), the discard priority of the packet is obtained (906) and a determination made as to whether the queue length plus one exceeds the relevant watermark (907).
If it does, the first packet in the queue is discarded (908) and a determination then made as to whether the queue length exceeds zero (909). If it does, the next packet is fetched (902) and the process continues as described above. Otherwise, if the queue length does not exceed zero, the process returns.
Assuming the queue length plus one does not exceed the relevant watermark (907), the selected packet will be transmitted (911) and the credit count will be decremented by an amount equal to the number of bytes in the transmitted packet (912). the queue length and the credit count will then be compared to a threshold - (such as "0")(913), and if appropriate amounts remain for both, a new packet will be fetched (902). Otherwise, the process will return .
Returning to Fig. 8, the process returns to determine whether any credit remains for this particular WO 92/01345 PCI'/US91/04444 ~56~043 18 queue group (804). Eventually, one of two events will occur. Either no packets will be found (806) and all of the queues in the group will have been scanned (811), or the credit for the group (804) will be completely 5 cieared. In either of these two events, the process will determine whether all queue groups have yet been scanned (812~. If not, the queue group identifier will be incremented (813) and the above described process will repeat for the next queue group. Otherwise, when all of 10 the queue groups have been examined, a determination will be made as to whether the credit counters for all queues have all become negative (814). If not, the process will begin again with the first queue group (802). Otherwise, appropriate steps will be taken (816 15 and 817) to reset the credit counters for all of the queue groups to an initial level, following which the above described procedures will again repeat.
The invention has been described for an application of the queueing discipline at the output of a packet 20 switch. The same techniques are equally applicable in the case of packet switches where the segregation and queueing on the basis of the traffic type is done at the input of the switch.
In addition, the methods described here are not 25 restricted to integrated networks that carry traffic from only CBO, voice, and data sources. For example, compressed video and image/facsimile traffic would be treated as separate traffic types if the network was to carry them. The bandwidth requirements of compressed 30 video fluctuate over time; when successive screens are not very different (for example, a slowly moving object) the bit rate is relatively small. On the other hand, rapidly moving objects or scene changes raise the instantaneous bandwidth requirements to higher levels.
35 In addition, packets from both video and image/facsimile WO 92/01345 PCr/l~ Og~4~ 3 sources can selectively be dropped at the cost of little or no degradation in the image quality.
What is claimed is:
Claims (31)
PROPERTY OF PRIVILEGE IS CLAIMED ARE DEFINED AS FOLLOWS:
1. A method of post-switching multiplexing fast packets for differing traffic types, comprising the steps of:
A) receiving first traffic type fast packets from a plurality of first sources;
prioritizing at least some of the first traffic type fast packets pursuant to a first prioritization method, to prioritize first traffic type fast packets for transmission;
B) receiving second traffic type fast packets from a plurality of second sources, which second traffic type is different from the first traffic type;
C) prioritizing at least some of the second traffic type fast packets pursuant to a second prioritization method, which second prioritization method is different from the first prioritization method is different from the first prioritization method, to prioritize second traffic type fast packets for transmission;
D) multiplexing an transmitting the first and second traffic type fast packets, wherein the step of multiplexing the first and second traffic type fast packets includes the step of multiplexing the first and second traffic type fast packets pursuant to a weighted round robin bandwidth allocation protocol.
A) receiving first traffic type fast packets from a plurality of first sources;
prioritizing at least some of the first traffic type fast packets pursuant to a first prioritization method, to prioritize first traffic type fast packets for transmission;
B) receiving second traffic type fast packets from a plurality of second sources, which second traffic type is different from the first traffic type;
C) prioritizing at least some of the second traffic type fast packets pursuant to a second prioritization method, which second prioritization method is different from the first prioritization method is different from the first prioritization method, to prioritize second traffic type fast packets for transmission;
D) multiplexing an transmitting the first and second traffic type fast packets, wherein the step of multiplexing the first and second traffic type fast packets includes the step of multiplexing the first and second traffic type fast packets pursuant to a weighted round robin bandwidth allocation protocol.
2. The method of claim 1 wherein the first prioritization method comprises a head-of-line prioritization method.
3. The method of claim 2 wherein the second prioritization method includes a packet discarding protocol.
4. The method of claim 3 wherein the packet discarding protocol includes the step of comparing a discard priority for a selected fast packet at the front of a selected fast packet queue with queue depth of a least one other fast packet queue.
5. A method of post-switching multiplesing fast packets for differing traffic types, comprising the steps of:
A) receiving differing traffic type fast packets;
B) prioritizing at least some of the fast packets for transmission as a function, at least in part, of a prioritization methjod, to provide prioritized fast packets;
C) selecting at least some of the prioritized fast packets for transmission as a function, at least in part, of a bandwidth allocation method, which bandwidth allocation method is different from the prioritization method, to provide multiplexed fast packets; and D) transmitting the multiplexed fast packets, wherein:
a) the first prioritization method comprises a head-of-line prioritization method; and b) the bandwidth allocation method comprises a weighted round robin bandwidth allocation mechanism.
A) receiving differing traffic type fast packets;
B) prioritizing at least some of the fast packets for transmission as a function, at least in part, of a prioritization methjod, to provide prioritized fast packets;
C) selecting at least some of the prioritized fast packets for transmission as a function, at least in part, of a bandwidth allocation method, which bandwidth allocation method is different from the prioritization method, to provide multiplexed fast packets; and D) transmitting the multiplexed fast packets, wherein:
a) the first prioritization method comprises a head-of-line prioritization method; and b) the bandwidth allocation method comprises a weighted round robin bandwidth allocation mechanism.
6. A method of post-switching multiplexing fast packets for differing traffic types, comprising the steps of:
A) receiving continuous bit-stream oriented fast packets from a plurality of first sources;
B) prioritizing at least some of the continuous bit-stream oriented fast packets pursuant to a first prioritization method, to prioritize the continuous bit-stream oriented fat packets for transmission;
C) receiving framed data fast packets from a plurality of second sources;
D) prioritizing a least some of the framed data fast packets pursuant to the first prioritization method, to prioritize the framed data fast packets for transmission;
E) multiplexing at least some of the prioritized comtinuous bit-stream oriented fast packets and framed data fast packets as a function, at least in part, of a bandwidth allocation method, to provide multiplexed fast packets.
A) receiving continuous bit-stream oriented fast packets from a plurality of first sources;
B) prioritizing at least some of the continuous bit-stream oriented fast packets pursuant to a first prioritization method, to prioritize the continuous bit-stream oriented fat packets for transmission;
C) receiving framed data fast packets from a plurality of second sources;
D) prioritizing a least some of the framed data fast packets pursuant to the first prioritization method, to prioritize the framed data fast packets for transmission;
E) multiplexing at least some of the prioritized comtinuous bit-stream oriented fast packets and framed data fast packets as a function, at least in part, of a bandwidth allocation method, to provide multiplexed fast packets.
7. The method of claim 6, wherein the bandwidth allocation method comprises a weighted round robin bandwidth allocation method.
8. A method of post-switching multiplexing fast packets for differing traffic types, comprising the steps of:
A) receiving continuous bit-stream oriented fast packets from a plurality of first sources;
B) prioritizing at least some of the continuous bit-stream oriented fast packets pursuant to a first prioritization method, to prioritize the continuous bit-stream oreinted fast packets for transmission;
C) receiving framed data fast packets from a plurality of second sources;
D) prioritizing at least some of the framed data fast packets pursuant to the first prioritization method, to prioritize the framed data fast packets for transmission;
E) multiplexing at least some of the prioritiezed continuous bit-stream oriented fast packets and framed data fast packets as a function, at least in part, of a bandwidth allocaiton method, to provide multiplexed fast packets, wherein:
a) the first method comprises a head-of-line prioritization method; and b) the bandwidth allocation method comprises a weighted round robin bandwidth allocation method.
A) receiving continuous bit-stream oriented fast packets from a plurality of first sources;
B) prioritizing at least some of the continuous bit-stream oriented fast packets pursuant to a first prioritization method, to prioritize the continuous bit-stream oreinted fast packets for transmission;
C) receiving framed data fast packets from a plurality of second sources;
D) prioritizing at least some of the framed data fast packets pursuant to the first prioritization method, to prioritize the framed data fast packets for transmission;
E) multiplexing at least some of the prioritiezed continuous bit-stream oriented fast packets and framed data fast packets as a function, at least in part, of a bandwidth allocaiton method, to provide multiplexed fast packets, wherein:
a) the first method comprises a head-of-line prioritization method; and b) the bandwidth allocation method comprises a weighted round robin bandwidth allocation method.
9. A mehtod of post-switching multiplexing fast packets for differing traffic types, comprising the steps of:
A) receiving continuous bit-stream oriented fast packets from a plurality of first sources;
B) prioritizing at least some of the continuous bit-stream oriented fast packets pursuant to a first prioritization method, to prioritize the continuous bit-stream oriented fast packets pursuant to a first prioritization method, to prioritize the continuous bit-stream oreinted fast packets for transmission;
C) receiving framed data fast packets from a plurality of second sources;
D) prioritizing at least some of the feramed data fast packets pursuant to the first prioritization method, to prioritize the framed data fast packets for transmission;
E) receiving digitized voice fast packets from a plurality of third sources;
F) discarding at least some of the digitized voice packets, from time to time, pursuant to a packet discarding protocol, to provide digitized voice fast packets for transmission;
G) multiplexing at least some of the prioritized continuous bit-stream oriented fast packets and framed data fast packets and digitized voice fast packets, as a funciton, at least in part, of a bandwidth allocation method, which bandwidth allocaiton method to provide multiplexed fast packets.
A) receiving continuous bit-stream oriented fast packets from a plurality of first sources;
B) prioritizing at least some of the continuous bit-stream oriented fast packets pursuant to a first prioritization method, to prioritize the continuous bit-stream oriented fast packets pursuant to a first prioritization method, to prioritize the continuous bit-stream oreinted fast packets for transmission;
C) receiving framed data fast packets from a plurality of second sources;
D) prioritizing at least some of the feramed data fast packets pursuant to the first prioritization method, to prioritize the framed data fast packets for transmission;
E) receiving digitized voice fast packets from a plurality of third sources;
F) discarding at least some of the digitized voice packets, from time to time, pursuant to a packet discarding protocol, to provide digitized voice fast packets for transmission;
G) multiplexing at least some of the prioritized continuous bit-stream oriented fast packets and framed data fast packets and digitized voice fast packets, as a funciton, at least in part, of a bandwidth allocation method, which bandwidth allocaiton method to provide multiplexed fast packets.
10. The method of claim 9, wherein the first method comprises a head-of-line prioritization method.
11. The method of claim 9, wherein the bandwidth allocation mehtod comprises a wighted round robin bandwidth allocation method.
12. The method of claim 9, wherein:
A) the first method comprises a head-of-line prioritization method; an B) the bandwidth allocaiton method comprises a weighted round robin bandwidth allocation method.
A) the first method comprises a head-of-line prioritization method; an B) the bandwidth allocaiton method comprises a weighted round robin bandwidth allocation method.
13. The method of claim 9 wherein at least some of the digitized voice fast packets include discard priority information, and wherein the packet discarding protocol utilizes the discard priority information to identify packets to discard.
14. A method of post-switching multiplexing fast packets for differing traffic types, comprising the steps of:
A) receiving continuous bit-stream oriented fast packets from a plurality of first sources;
B) storing at least some of the continuous bit-stream oriented fast packets in at least a first queue and a second queue;
C) prioritizing at least some of the continuous bit-stream oreinted fast packets as stored in teh queues pursuant to a first prioritization method, to prioritize the continuous bit-strem oriented fast packets for transmission;
D) receiving framed data fast packets from a plurality of second sources;
E) storing at least some of the frame data fast packets in at least a third queue and a fourth queue;
F) prioritizing at least some of the framed data fast packets as stored in the third and fourth queues prusuant to the second prioritization method, to prioritize the framed data fast packets for transmission;
G) receiving digitized voice fast packets from a plurality of third sources;
H) discarding at least some of the digitized voice packts, from time to time, pursuant to a packet discarding protocol, to provide digitized voice fast packets for transmission;
l) multiplexing at least some of the prioritized continuous bit-stream oriented fast packets and framed data fast packets and digitized voice fast packets, as a funciton, at least in part, of a bandwidth allocaiton method, which bandwidth allocaiton method to provide multiplexed fast packets.
A) receiving continuous bit-stream oriented fast packets from a plurality of first sources;
B) storing at least some of the continuous bit-stream oriented fast packets in at least a first queue and a second queue;
C) prioritizing at least some of the continuous bit-stream oreinted fast packets as stored in teh queues pursuant to a first prioritization method, to prioritize the continuous bit-strem oriented fast packets for transmission;
D) receiving framed data fast packets from a plurality of second sources;
E) storing at least some of the frame data fast packets in at least a third queue and a fourth queue;
F) prioritizing at least some of the framed data fast packets as stored in the third and fourth queues prusuant to the second prioritization method, to prioritize the framed data fast packets for transmission;
G) receiving digitized voice fast packets from a plurality of third sources;
H) discarding at least some of the digitized voice packts, from time to time, pursuant to a packet discarding protocol, to provide digitized voice fast packets for transmission;
l) multiplexing at least some of the prioritized continuous bit-stream oriented fast packets and framed data fast packets and digitized voice fast packets, as a funciton, at least in part, of a bandwidth allocaiton method, which bandwidth allocaiton method to provide multiplexed fast packets.
15. The method of claim 14, wherein the continuous bit-stream oriented fast packets are stored in the first queue and the second queue as a function in part of the packetization time of the CBO sources.
16. The method of claim 15, wherein the continuous bit-stream oriented fast packets from sources having a relatively small packetization time are stored in the first queue, and the fast packets having a relatively large packetization time are stored in the second queue.
17. The method of claim 16, wherein the step of storing the continuous bit-stream packets includes toring fast packets from sources with a relatively medium packetization times in additional queues that are served via HOLP
with the first and second queues.
with the first and second queues.
18. The method of claim 14, wherein the framed data fast packets are stored int he third queue and the fourth queue as a funciton in part of the burst size of the framed data sources.
19. The method of claim 18, wherein the framed data fast packets from sources having a relatively small burst size are stored in the third queue, and the fast packets having a relatively large burst size are stored in the fourth queue.
20. The method of claim 19, wherein the step of storing the framed data packets includes storing rfast packets from sources with relatively medium burst sizes in aditional queues that are served via HOLP with the third and fourth queues.
21. A method of post-switching multiplexing fast packets for differing traffic types, wherein fast packets for each traffic type are queued in a plurality of queues, such that queues that contain fast packets for a particular type are grouped together in a discrete group, comprising the steps of:
A) selecting a first group fo queues;
B) determining whether a credit count associated with the first group of queues has any remaining credit contained therein;
C) when the credit count has credit at least a predetermined amount of remaining therein, determining whether a queue contained within the group has a fast apcket queued therein;
D) when the queue has a fast packet queued therein, removing the fast packet from the queue;
E) when the fast packet queued in the queue is removed by transmission of the fast packet, altering the credit count.
A) selecting a first group fo queues;
B) determining whether a credit count associated with the first group of queues has any remaining credit contained therein;
C) when the credit count has credit at least a predetermined amount of remaining therein, determining whether a queue contained within the group has a fast apcket queued therein;
D) when the queue has a fast packet queued therein, removing the fast packet from the queue;
E) when the fast packet queued in the queue is removed by transmission of the fast packet, altering the credit count.
22. The method of claim 21, wherein the step of altering the credit count includes the step of altering the credit count by an amount related to at least the length of the fast packet.
23. The method of claim 22, wherein step C further includes teh step of:
C1) when the credit count ahs less than a predetermined amount of credit remaining therein, repeating from step A with a second group of queues.
C1) when the credit count ahs less than a predetermined amount of credit remaining therein, repeating from step A with a second group of queues.
24. The method of claim 22, and further including the step of:
F) repeating from step A.
F) repeating from step A.
25. The method of claim 24, wherein step C further includes teh step of:
C1) when the credit count ahs less than a predetermined amount of credit remining therein, repeating from step A with a second group of queues.
C1) when the credit count ahs less than a predetermined amount of credit remining therein, repeating from step A with a second group of queues.
26. The method of claim 25, and further including the step of:
G) determing whether any of the groups of queues have any credit remaining in their respective credit counts;
H) when no credit remains with any of the groups of queues, establishing at least a predetermined credit count for a least some of the groups of queues.
G) determing whether any of the groups of queues have any credit remaining in their respective credit counts;
H) when no credit remains with any of the groups of queues, establishing at least a predetermined credit count for a least some of the groups of queues.
27. The method of claim 25, wherein a predetermined credit count is established for each of the groups of queues.
28. The method of claim 27, wherein the predetermined credit count is different for at least some of the groups of queues.
29. The method of claim 22, wherein the fast packets are comprised of bytes and the step of altering the credit count by an amount related to at least the length of the fast packet includes reducing the credit count by an amount proportional to the number of bytes in the removed fast packet.
30. A method of post-switching multiplexing fast packets for differing traffic types, comprising the steps of:
A) receiving continuous bit-stream oriented fast packets from a plurality of first sources;
B) prioritizing at least some of the continuous bit-stream oriented fast packets pursuant to a first prioritization method, to prioritize the continuous bit-stream oriented fast packets for transmission;
C) receiving framed data fast packets from a plurality of second sources;
D) prioritizing at least some of the framed data fast packets pursuant to the first prioritization method, to prioritize the framed data fast packets for transmission;
E) multiplexing at least some of the prioritized continuous bit-stream oriented fast packets and framed data fast packets as a function, at least in part, of a bandwidth allocation method, to provide multiplexed fast packets, wherein the bandwidth allocation method comprises a weighted round robin bandwidth allocation method.
A) receiving continuous bit-stream oriented fast packets from a plurality of first sources;
B) prioritizing at least some of the continuous bit-stream oriented fast packets pursuant to a first prioritization method, to prioritize the continuous bit-stream oriented fast packets for transmission;
C) receiving framed data fast packets from a plurality of second sources;
D) prioritizing at least some of the framed data fast packets pursuant to the first prioritization method, to prioritize the framed data fast packets for transmission;
E) multiplexing at least some of the prioritized continuous bit-stream oriented fast packets and framed data fast packets as a function, at least in part, of a bandwidth allocation method, to provide multiplexed fast packets, wherein the bandwidth allocation method comprises a weighted round robin bandwidth allocation method.
31. The method of claim 30, wherein the first method comprises a head-of-line prioritization method.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US07/551,712 US5231633A (en) | 1990-07-11 | 1990-07-11 | Method for prioritizing, selectively discarding, and multiplexing differing traffic type fast packets |
| US551,712 | 1990-07-11 |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CA2065043A1 CA2065043A1 (en) | 1992-01-12 |
| CA2065043C true CA2065043C (en) | 1996-07-02 |
Family
ID=24202374
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CA002065043A Expired - Lifetime CA2065043C (en) | 1990-07-11 | 1991-06-24 | Method for prioritizing selectively discarding and multiplexing differing traffic type fast packets |
Country Status (7)
| Country | Link |
|---|---|
| US (1) | US5231633A (en) |
| EP (1) | EP0494284B1 (en) |
| JP (1) | JP2593120B2 (en) |
| AU (1) | AU629757B2 (en) |
| CA (1) | CA2065043C (en) |
| DE (1) | DE69130286T2 (en) |
| WO (1) | WO1992001345A1 (en) |
Families Citing this family (304)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5280483A (en) * | 1990-08-09 | 1994-01-18 | Fujitsu Limited | Traffic control system for asynchronous transfer mode exchange |
| DE59206183D1 (en) * | 1992-01-30 | 1996-06-05 | Siemens Ag | Method for limiting the rate of message cells of a virtual connection |
| JPH05292114A (en) * | 1992-04-09 | 1993-11-05 | Fujitsu Ltd | Communication path setting device and its method |
| JP2682770B2 (en) * | 1992-05-15 | 1997-11-26 | 富士通株式会社 | CPU control method for virtual computer system |
| CA2104753C (en) * | 1992-10-29 | 1999-02-16 | Kotikalapudi Sriram | Bandwidth allocation, transmission scheduling, and congestion avoidance in broadband atm networks |
| US5691975A (en) * | 1993-03-12 | 1997-11-25 | Fujitsu Limited | Method and apparatus for control in ATM |
| CA2164597A1 (en) * | 1993-06-07 | 1994-12-22 | Duncan Hartley Tate | Communication system |
| JP3507524B2 (en) * | 1993-06-21 | 2004-03-15 | 富士通株式会社 | Data processing system with event report word processing function |
| US5696764A (en) * | 1993-07-21 | 1997-12-09 | Fujitsu Limited | ATM exchange for monitoring congestion and allocating and transmitting bandwidth-guaranteed and non-bandwidth-guaranteed connection calls |
| JP3044983B2 (en) * | 1993-08-25 | 2000-05-22 | 株式会社日立製作所 | Cell switching method for ATM switching system |
| US5432790A (en) * | 1993-09-17 | 1995-07-11 | Motorola, Inc. | Method for allocating internodal link bandwidth in a packet oriented communication network to guarantee delay quality-of-service |
| US5408472A (en) * | 1993-09-20 | 1995-04-18 | Motorola, Inc. | Device and method for cell processing in cell relay nodes |
| US5446736A (en) * | 1993-10-07 | 1995-08-29 | Ast Research, Inc. | Method and apparatus for connecting a node to a wireless network using a standard protocol |
| US5528513A (en) * | 1993-11-04 | 1996-06-18 | Digital Equipment Corp. | Scheduling and admission control policy for a continuous media server |
| FI94814C (en) * | 1993-11-30 | 1995-10-25 | Nokia Telecommunications Oy | Procedure for managing congestion situations in a frame switching network and a subscriber node in a frame switching network |
| GB9326276D0 (en) * | 1993-12-23 | 1994-02-23 | Newbridge Network Corp | Frame relay interface |
| JPH07183888A (en) * | 1993-12-24 | 1995-07-21 | Fujitsu Ltd | Atm multiplexing control system |
| US5991301A (en) * | 1994-05-05 | 1999-11-23 | Sprint Communications Co. L.P. | Broadband telecommunications system |
| US6633561B2 (en) | 1994-05-05 | 2003-10-14 | Sprint Communications Company, L.P. | Method, system and apparatus for telecommunications control |
| US6314103B1 (en) | 1994-05-05 | 2001-11-06 | Sprint Communications Company, L.P. | System and method for allocating bandwidth for a call |
| US5926482A (en) | 1994-05-05 | 1999-07-20 | Sprint Communications Co. L.P. | Telecommunications apparatus, system, and method with an enhanced signal transfer point |
| JPH10500542A (en) | 1994-05-05 | 1998-01-13 | スプリント コミュニケーションズ カンパニー，エル．ピー． | Method, system (system) and device for telecommunication control |
| US6023474A (en) * | 1996-11-22 | 2000-02-08 | Sprint Communications C.O.L.P. | Broadband telecommunications system interface |
| US5920562A (en) * | 1996-11-22 | 1999-07-06 | Sprint Communications Co. L.P. | Systems and methods for providing enhanced services for telecommunication call |
| US6430195B1 (en) | 1994-05-05 | 2002-08-06 | Sprint Communications Company L.P. | Broadband telecommunications system interface |
| US6031840A (en) * | 1995-12-07 | 2000-02-29 | Sprint Communications Co. L.P. | Telecommunications system |
| US6172977B1 (en) * | 1994-05-05 | 2001-01-09 | Sprint Communications Company, L. P. | ATM direct access line system |
| US6181703B1 (en) * | 1995-09-08 | 2001-01-30 | Sprint Communications Company L. P. | System for managing telecommunications |
| FI98774C (en) * | 1994-05-24 | 1997-08-11 | Nokia Telecommunications Oy | Method and apparatus for prioritizing traffic in an ATM network |
| US5487061A (en) * | 1994-06-27 | 1996-01-23 | Loral Fairchild Corporation | System and method for providing multiple loss and service priorities |
| US5553061A (en) * | 1994-06-27 | 1996-09-03 | Loral Fairchild Corporation | Packet processor having service priority and loss priority features |
| DE69430627T2 (en) * | 1994-06-28 | 2002-10-17 | Hewlett Packard Co | Method and device for planning cell transmission of virtual channels with guaranteed bandwidth |
| US5923657A (en) * | 1994-08-23 | 1999-07-13 | Hitachi, Ltd. | ATM switching system and cell control method |
| SE9403101L (en) * | 1994-09-16 | 1996-03-17 | Ericsson Telefon Ab L M | Procedure for customizing data flows |
| US5533020A (en) * | 1994-10-31 | 1996-07-02 | International Business Machines Corporation | ATM cell scheduler |
| US5881313A (en) * | 1994-11-07 | 1999-03-09 | Digital Equipment Corporation | Arbitration system based on requester class and relative priority including transmit descriptor valid bit for a shared resource having multiple requesters |
| EP0714192A1 (en) * | 1994-11-24 | 1996-05-29 | International Business Machines Corporation | Method for preempting connections in high speed packet switching networks |
| JPH08163145A (en) * | 1994-12-08 | 1996-06-21 | Nec Corp | Traffic shaping method for atm exchange and device therefor |
| EP0717532A1 (en) * | 1994-12-13 | 1996-06-19 | International Business Machines Corporation | Dynamic fair queuing to support best effort traffic in an ATM network |
| US5793978A (en) * | 1994-12-29 | 1998-08-11 | Cisco Technology, Inc. | System for routing packets by separating packets in to broadcast packets and non-broadcast packets and allocating a selected communication bandwidth to the broadcast packets |
| US5606553A (en) * | 1995-02-28 | 1997-02-25 | Christie; Joseph M. | Cell processing for voice transmission |
| JPH08307420A (en) * | 1995-03-03 | 1996-11-22 | Fujitsu Ltd | Congestion control system for cell exchange |
| US5675573A (en) * | 1995-03-22 | 1997-10-07 | Lucent Technologies Inc. | Delay-minimizing system with guaranteed bandwidth delivery for real-time traffic |
| US7224819B2 (en) * | 1995-05-08 | 2007-05-29 | Digimarc Corporation | Integrating digital watermarks in multimedia content |
| GB9509921D0 (en) * | 1995-05-17 | 1995-07-12 | Roke Manor Research | Improvements in or relating to mobile radio systems |
| US6522635B1 (en) * | 1995-06-15 | 2003-02-18 | Mobile Satellite Ventures, Lp | Communication protocol for satellite data processing |
| US5889764A (en) * | 1995-08-31 | 1999-03-30 | Intel Corporation | Low-latency multi-party audio chat |
| US5784568A (en) * | 1995-08-31 | 1998-07-21 | Intel Corporation | Multi-party audio chat system which allows individual user utterances to be staged separately to render received utterances in order |
| US5764641A (en) * | 1995-09-08 | 1998-06-09 | Cisco Systems, Inc. | Early and integrated tail packet discard system |
| GB9520807D0 (en) * | 1995-10-11 | 1995-12-13 | Newbridge Networks Corp | Fair queue servicing using dynamic weights |
| US5831985A (en) * | 1995-11-09 | 1998-11-03 | Emc Corporation | Method and apparatus for controlling concurrent data transmission from multiple sources in a channel communication system |
| US6327246B1 (en) | 1995-11-29 | 2001-12-04 | Ahead Communications Systems, Inc. | Controlled available bit rate service in an ATM switch |
| GB2307823B (en) * | 1995-11-29 | 2000-04-12 | Gen Datacomm Adv Res | Available bit rate services in ATM networks |
| DE19548985A1 (en) * | 1995-12-28 | 1997-07-03 | Siemens Ag | Method for prioritizing cell streams in systems that transmit information according to an asynchronous transfer mode (ATM) |
| AU2257097A (en) * | 1996-02-02 | 1997-08-22 | Sprint Communications Company, L.P. | Atm gateway system |
| US5870550A (en) * | 1996-02-26 | 1999-02-09 | Network Engineering Software | Web server employing multi-homed, moldular framework |
| US8117298B1 (en) | 1996-02-26 | 2012-02-14 | Graphon Corporation | Multi-homed web server |
| US6011587A (en) * | 1996-03-07 | 2000-01-04 | Kokusai Denshin Denwa Kabushiki Kaisha | Packet video bitrate conversion system |
| US5870629A (en) * | 1996-03-21 | 1999-02-09 | Bay Networks, Inc. | System for servicing plurality of queues responsive to queue service policy on a service sequence ordered to provide uniform and minimal queue interservice times |
| US6219728B1 (en) | 1996-04-22 | 2001-04-17 | Nortel Networks Limited | Method and apparatus for allocating shared memory resources among a plurality of queues each having a threshold value therefor |
| US6308148B1 (en) | 1996-05-28 | 2001-10-23 | Cisco Technology, Inc. | Network flow data export |
| US5940393A (en) * | 1996-05-28 | 1999-08-17 | Sprint Communications Co. L.P. | Telecommunications system with a connection processing system |
| JPH104431A (en) * | 1996-06-17 | 1998-01-06 | Fujitsu Ltd | Scheduling device and method |
| US5926459A (en) * | 1996-06-27 | 1999-07-20 | Xerox Corporation | Rate shaping in per-flow queued routing mechanisms for available bit rate service |
| US6064677A (en) * | 1996-06-27 | 2000-05-16 | Xerox Corporation | Multiple rate sensitive priority queues for reducing relative data transport unit delay variations in time multiplexed outputs from output queued routing mechanisms |
| US6073177A (en) * | 1997-08-05 | 2000-06-06 | Sterling Software, Inc. | Dynamic method for connecting a client to a server application |
| US7181525B1 (en) | 1996-08-05 | 2007-02-20 | Computer Associates Think, Inc. | Dynamic method for connecting a client to a server application |
| JP3435293B2 (en) * | 1996-09-10 | 2003-08-11 | 株式会社東芝 | Packet scheduling apparatus and packet transfer method |
| FI103455B (en) | 1996-10-08 | 1999-06-30 | Nokia Telecommunications Oy | Packet router |
| EP0843499A3 (en) * | 1996-11-19 | 1999-01-20 | Italtel s.p.a. | Method and device for the management of resources in ATM technique for weighted fair queuing (WFQ) applications |
| US6115380A (en) * | 1996-11-22 | 2000-09-05 | Sprint Communications Co., L.P. | Broadband telecommunications system |
| US6002689A (en) * | 1996-11-22 | 1999-12-14 | Sprint Communications Co. L.P. | System and method for interfacing a local communication device |
| US6014378A (en) * | 1996-11-22 | 2000-01-11 | Sprint Communications Company, L.P. | Telecommunications tandem system for circuit-based traffic |
| CA2271926C (en) | 1996-11-22 | 2005-10-11 | Sprint Communications Company, L.P. | System and method for transporting a call in a telecommunication network |
| US6208655B1 (en) * | 1996-11-27 | 2001-03-27 | Sony Europa, B.V., | Method and apparatus for serving data |
| US6124878A (en) * | 1996-12-20 | 2000-09-26 | Time Warner Cable, A Division Of Time Warner Enterainment Company, L.P. | Optimum bandwidth utilization in a shared cable system data channel |
| US6144637A (en) * | 1996-12-20 | 2000-11-07 | Cisco Technology, Inc. | Data communications |
| US5926458A (en) * | 1997-01-31 | 1999-07-20 | Bay Networks | Method and apparatus for servicing multiple queues |
| US5953335A (en) * | 1997-02-14 | 1999-09-14 | Advanced Micro Devices, Inc. | Method and apparatus for selectively discarding packets for blocked output queues in the network switch |
| KR19980073015A (en) * | 1997-03-11 | 1998-11-05 | 김광호 | Video conferencing system |
| CA2285245A1 (en) * | 1997-04-04 | 1998-10-15 | Mahesh N. Ganmukhi | High speed packet scheduling method and apparatus |
| US6088331A (en) * | 1997-04-07 | 2000-07-11 | Nortel Networks Corporation | Parallel service-rate controllers in ATM switches |
| US6067299A (en) * | 1997-04-16 | 2000-05-23 | Sprint Communications Company, L.P. | Communications system for providing ATM connections and echo cancellation |
| US6041059A (en) * | 1997-04-25 | 2000-03-21 | Mmc Networks, Inc. | Time-wheel ATM cell scheduling |
| US6014367A (en) * | 1997-04-25 | 2000-01-11 | Mmc Networks, Inc | Method for weighted fair queuing for ATM cell scheduling |
| US6144638A (en) * | 1997-05-09 | 2000-11-07 | Bbn Corporation | Multi-tenant unit |
| US6137800A (en) * | 1997-05-09 | 2000-10-24 | Sprint Communications Company, L. P. | System and method for connecting a call |
| US6704327B1 (en) | 1997-05-09 | 2004-03-09 | Sprint Communications Company, L.P. | System and method for connecting a call |
| US6178170B1 (en) | 1997-05-13 | 2001-01-23 | Sprint Communications Company, L. P. | System and method for transporting a call |
| DE69733129T2 (en) * | 1997-06-20 | 2006-03-09 | Alcatel | Method and device for transmitting data packets with priorities |
| US6052738A (en) * | 1997-06-30 | 2000-04-18 | Sun Microsystems, Inc. | Method and apparatus in a packet routing switch for controlling access at different data rates to a shared memory |
| US6044418A (en) * | 1997-06-30 | 2000-03-28 | Sun Microsystems, Inc. | Method and apparatus for dynamically resizing queues utilizing programmable partition pointers |
| US6119196A (en) * | 1997-06-30 | 2000-09-12 | Sun Microsystems, Inc. | System having multiple arbitrating levels for arbitrating access to a shared memory by network ports operating at different data rates |
| US5920566A (en) * | 1997-06-30 | 1999-07-06 | Sun Microsystems, Inc. | Routing in a multi-layer distributed network element |
| US6049528A (en) * | 1997-06-30 | 2000-04-11 | Sun Microsystems, Inc. | Trunking ethernet-compatible networks |
| US6246680B1 (en) | 1997-06-30 | 2001-06-12 | Sun Microsystems, Inc. | Highly integrated multi-layer switch element architecture |
| US6081512A (en) * | 1997-06-30 | 2000-06-27 | Sun Microsystems, Inc. | Spanning tree support in a high performance network device |
| US6128666A (en) * | 1997-06-30 | 2000-10-03 | Sun Microsystems, Inc. | Distributed VLAN mechanism for packet field replacement in a multi-layered switched network element using a control field/signal for indicating modification of a packet with a database search engine |
| US5938736A (en) * | 1997-06-30 | 1999-08-17 | Sun Microsystems, Inc. | Search engine architecture for a high performance multi-layer switch element |
| US6088356A (en) * | 1997-06-30 | 2000-07-11 | Sun Microsystems, Inc. | System and method for a multi-layer network element |
| US6044087A (en) * | 1997-06-30 | 2000-03-28 | Sun Microsystems, Inc. | Interface for a highly integrated ethernet network element |
| US6094435A (en) * | 1997-06-30 | 2000-07-25 | Sun Microsystems, Inc. | System and method for a quality of service in a multi-layer network element |
| US6016310A (en) * | 1997-06-30 | 2000-01-18 | Sun Microsystems, Inc. | Trunking support in a high performance network device |
| US6014380A (en) * | 1997-06-30 | 2000-01-11 | Sun Microsystems, Inc. | Mechanism for packet field replacement in a multi-layer distributed network element |
| US6081522A (en) * | 1997-06-30 | 2000-06-27 | Sun Microsystems, Inc. | System and method for a multi-layer network element |
| JP3216582B2 (en) * | 1997-07-18 | 2001-10-09 | 日本電気株式会社 | Data output control method |
| NO309918B1 (en) * | 1997-08-18 | 2001-04-17 | Ericsson Telefon Ab L M | Procedure related to a data communication system |
| DE19737852C2 (en) * | 1997-08-29 | 2001-06-07 | Siemens Ag | Method and circuit arrangement for transmitting message units in message streams of different priority |
| US6473425B1 (en) * | 1997-10-02 | 2002-10-29 | Sun Microsystems, Inc. | Mechanism for dispatching packets via a telecommunications network |
| JP2959540B2 (en) * | 1997-10-03 | 1999-10-06 | 日本電気株式会社 | Node device |
| US6108697A (en) * | 1997-10-06 | 2000-08-22 | Powerquest Corporation | One-to-many disk imaging transfer over a network |
| US6202094B1 (en) * | 1997-11-07 | 2001-03-13 | International Business Machines Corporation | Adding links simultaneously to a multilink bundle using bandwidth allocation protocol |
| US7260610B2 (en) * | 1998-02-10 | 2007-08-21 | Gateway Inc. | Convergence events notification system |
| US6563918B1 (en) | 1998-02-20 | 2003-05-13 | Sprint Communications Company, LP | Telecommunications system architecture for connecting a call |
| US6922409B1 (en) | 1998-02-20 | 2005-07-26 | Sprint Communications Company L.P. | System and method for allocating bandwidth for a call |
| US6574222B1 (en) | 1998-02-20 | 2003-06-03 | Sprint Communications Company, L.P. | System and method for allocating bandwidth for a call |
| US7079534B1 (en) | 1998-02-20 | 2006-07-18 | Sprint Communications Company L.P. | System and method for allocating bandwidth for a call |
| US6483837B1 (en) | 1998-02-20 | 2002-11-19 | Sprint Communications Company L.P. | System and method for connecting a call with an interworking system |
| US6570876B1 (en) * | 1998-04-01 | 2003-05-27 | Hitachi, Ltd. | Packet switch and switching method for switching variable length packets |
| US6546022B1 (en) | 1998-04-03 | 2003-04-08 | Sprint Communications Company, L.P. | Method, system and apparatus for processing information in a telecommunications system |
| US6160871A (en) | 1998-04-10 | 2000-12-12 | Sprint Communications Company, L.P. | Communications test system |
| US6366589B1 (en) * | 1998-04-27 | 2002-04-02 | 3Com Corporation | Multiple parallel asymmetric interfaces with reversed asymmetric links |
| US6067301A (en) * | 1998-05-29 | 2000-05-23 | Cabletron Systems, Inc. | Method and apparatus for forwarding packets from a plurality of contending queues to an output |
| US6370121B1 (en) | 1998-06-29 | 2002-04-09 | Cisco Technology, Inc. | Method and system for shortcut trunking of LAN bridges |
| US6377547B1 (en) * | 1998-06-30 | 2002-04-23 | At&T Corp. | Quality of service-oriented cell disc discarding |
| IL125271A0 (en) * | 1998-07-08 | 1999-03-12 | Galileo Technology Ltd | Head of line blocking |
| CA2243676A1 (en) * | 1998-07-22 | 2000-01-22 | Newbridge Networks Corporation | Frame reference load model generator |
| US6246691B1 (en) | 1998-08-14 | 2001-06-12 | Siemens Aktiengesellschaft | Method and circuit configuration for the transmission of message units in message streams of different priority |
| US6421720B2 (en) * | 1998-10-28 | 2002-07-16 | Cisco Technology, Inc. | Codec-independent technique for modulating bandwidth in packet network |
| KR100279622B1 (en) * | 1998-10-28 | 2001-02-01 | 구자홍 | Method and apparatus for controlling voice packet transmission |
| US6614781B1 (en) | 1998-11-20 | 2003-09-02 | Level 3 Communications, Inc. | Voice over data telecommunications network architecture |
| US6442169B1 (en) | 1998-11-20 | 2002-08-27 | Level 3 Communications, Inc. | System and method for bypassing data from egress facilities |
| US6229813B1 (en) * | 1998-11-25 | 2001-05-08 | Alcatel Canada Inc. | Pointer system for queue size control in a multi-task processing application |
| US6885657B1 (en) * | 1998-11-30 | 2005-04-26 | Broadcom Corporation | Network telephony system |
| US6490298B1 (en) | 1999-02-26 | 2002-12-03 | Harmonic Inc. | Apparatus and methods of multiplexing data to a communication channel |
| US6606301B1 (en) * | 1999-03-01 | 2003-08-12 | Sun Microsystems, Inc. | Method and apparatus for early random discard of packets |
| US6539026B1 (en) | 1999-03-15 | 2003-03-25 | Cisco Technology, Inc. | Apparatus and method for delay management in a data communications network |
| US6539024B1 (en) * | 1999-03-26 | 2003-03-25 | Alcatel Canada Inc. | Method and apparatus for data buffer management in a communications switch |
| US7392279B1 (en) * | 1999-03-26 | 2008-06-24 | Cisco Technology, Inc. | Network traffic shaping using time-based queues |
| US7103068B1 (en) | 1999-05-04 | 2006-09-05 | Sprint Communication Company L.P. | System and method for configuring bandwidth transmission rates for call connections |
| JP3438651B2 (en) * | 1999-05-31 | 2003-08-18 | 日本電気株式会社 | Packet multiplexer |
| US6675220B1 (en) * | 1999-08-11 | 2004-01-06 | Cisco Technology, Inc. | Techniques for the hardware implementation of random early detection mechanisms |
| JP3683133B2 (en) | 1999-08-17 | 2005-08-17 | 日本電気株式会社 | Packet scheduling device |
| US6618380B1 (en) * | 1999-09-09 | 2003-09-09 | Lucent Technologies Inc. | Priority queueing |
| US6674718B1 (en) | 2000-04-11 | 2004-01-06 | International Business Machines Corporation | Unified method and system for scheduling and discarding packets in computer networks |
| US6771652B1 (en) | 1999-11-23 | 2004-08-03 | International Business Machines Corporation | Method and system for controlling transmission of packets in computer networks |
| US6657960B1 (en) | 1999-11-23 | 2003-12-02 | International Business Machines Corporation | Method and system for providing differentiated services in computer networks |
| US6625122B1 (en) | 1999-11-24 | 2003-09-23 | Applied Micro Circuits Corporation | Selection of data for network transmission |
| JP2001156785A (en) * | 1999-11-26 | 2001-06-08 | Nec Corp | Cell abort avoidance system and method in buffer |
| US6577596B1 (en) * | 1999-11-30 | 2003-06-10 | Telefonaktiebolaget Ln Ericsson (Publ) | Method and apparatus for packet delay reduction using scheduling and header compression |
| US6704314B1 (en) | 1999-12-15 | 2004-03-09 | Sprint Communications Company, L.P. | Method and apparatus to control cell substitution |
| US20010044107A1 (en) * | 1999-12-28 | 2001-11-22 | Zarling David A. | Recombinase mediated gene chip detection |
| US6856613B1 (en) * | 1999-12-30 | 2005-02-15 | Cisco Technology, Inc. | Method and apparatus for throttling audio packets according to gateway processing capacity |
| US6735214B1 (en) * | 2000-01-12 | 2004-05-11 | New Jersey Institute Of Technology | Method and system for a hierarchical traffic shaper |
| US6882623B1 (en) | 2000-02-08 | 2005-04-19 | Native Networks Technologies Ltd. | Multi-level scheduling method for multiplexing packets in a communications network |
| US6813243B1 (en) * | 2000-02-14 | 2004-11-02 | Cisco Technology, Inc. | High-speed hardware implementation of red congestion control algorithm |
| JP3891755B2 (en) | 2000-03-27 | 2007-03-14 | 沖電気工業株式会社 | Packet receiver |
| US6769043B1 (en) * | 2000-04-11 | 2004-07-27 | Cisco Technology, Inc. | Ensuring fair access to upstream trunk bandwidth in ATM subtended configurations |
| US7324635B2 (en) | 2000-05-04 | 2008-01-29 | Telemaze Llc | Branch calling and caller ID based call routing telephone features |
| US7151773B1 (en) | 2000-05-05 | 2006-12-19 | Fujitsu Limited | System and method for connectionless/connection oriented signal transport |
| US7075927B2 (en) * | 2000-05-05 | 2006-07-11 | Fujitsu Limited | Method and system for quality of service (QoS) support in a packet-switched network |
| US7133403B1 (en) | 2000-05-05 | 2006-11-07 | Fujitsu Limited | Transport network and method |
| US7385917B1 (en) | 2000-05-05 | 2008-06-10 | Fujitsu Limited | Method and system for providing a protection path for connectionless signals in a telecommunications network |
| US7058730B2 (en) * | 2000-05-05 | 2006-06-06 | Fujitsu Limited | Unique address space and method for a transport network |
| US6775229B1 (en) | 2000-05-05 | 2004-08-10 | Fujitsu Network Communications, Inc. | Method and system for providing a protection path for connection-oriented signals in a telecommunications network |
| US7173912B2 (en) * | 2000-05-05 | 2007-02-06 | Fujitsu Limited | Method and system for modeling and advertising asymmetric topology of a node in a transport network |
| US6693909B1 (en) | 2000-05-05 | 2004-02-17 | Fujitsu Network Communications, Inc. | Method and system for transporting traffic in a packet-switched network |
| US7047176B2 (en) * | 2000-05-05 | 2006-05-16 | Fujitsu Limited | Method and system for hardware simulation |
| US6515966B1 (en) | 2000-05-05 | 2003-02-04 | Fujitsu Network Communications, Inc. | System and method for application object transport |
| US6775293B1 (en) * | 2000-06-30 | 2004-08-10 | Alcatel Canada Inc. | Method and apparatus for monitoring buffer contents in a data communication system |
| US6404772B1 (en) * | 2000-07-27 | 2002-06-11 | Symbol Technologies, Inc. | Voice and data wireless communications network and method |
| US6990113B1 (en) * | 2000-09-08 | 2006-01-24 | Mitsubishi Electric Research Labs., Inc. | Adaptive-weighted packet scheduler for supporting premium service in a communications network |
| US7142514B2 (en) * | 2000-09-29 | 2006-11-28 | Zarlink Semiconductor V.N. Inc. | Bandwidth sharing using emulated weighted fair queuing |
| US20020075886A1 (en) * | 2000-10-04 | 2002-06-20 | Tagore-Brage Jens P. | Data communication network switching unit having a systolic ring structure |
| US7237011B1 (en) * | 2000-11-01 | 2007-06-26 | Sun Microsystems, Inc. | System and method for a priority messaging protocol for a shared display device |
| US7596139B2 (en) | 2000-11-17 | 2009-09-29 | Foundry Networks, Inc. | Backplane interface adapter with error control and redundant fabric |
| US7236490B2 (en) | 2000-11-17 | 2007-06-26 | Foundry Networks, Inc. | Backplane interface adapter |
| US20020126673A1 (en) * | 2001-01-12 | 2002-09-12 | Nirav Dagli | Shared memory |
| US20020095684A1 (en) * | 2001-01-12 | 2002-07-18 | St. John James T. | Methods, systems and computer program products for bandwidth allocation based on throughput guarantees |
| US6944171B2 (en) * | 2001-03-12 | 2005-09-13 | Switchcore, Ab | Scheduler method and device in a switch |
| US6944168B2 (en) * | 2001-05-04 | 2005-09-13 | Slt Logic Llc | System and method for providing transformation of multi-protocol packets in a data stream |
| US6904057B2 (en) * | 2001-05-04 | 2005-06-07 | Slt Logic Llc | Method and apparatus for providing multi-protocol, multi-stage, real-time frame classification |
| US6901052B2 (en) * | 2001-05-04 | 2005-05-31 | Slt Logic Llc | System and method for policing multiple data flows and multi-protocol data flows |
| US7042848B2 (en) * | 2001-05-04 | 2006-05-09 | Slt Logic Llc | System and method for hierarchical policing of flows and subflows of a data stream |
| US7042841B2 (en) * | 2001-07-16 | 2006-05-09 | International Business Machines Corporation | Controlling network congestion using a biased packet discard policy for congestion control and encoded session packets: methods, systems, and program products |
| US7139280B2 (en) * | 2001-07-30 | 2006-11-21 | Yishay Mansour | Buffer management policy for shared memory switches |
| US7027443B2 (en) * | 2001-08-23 | 2006-04-11 | Pmc-Sierra Ltd. | Reassembly engines for multilink applications |
| ATE301902T1 (en) * | 2001-08-29 | 2005-08-15 | Cit Alcatel | ROUTERS |
| ATE289143T1 (en) | 2001-08-29 | 2005-02-15 | Cit Alcatel | ROUTERS |
| US7099275B2 (en) * | 2001-09-21 | 2006-08-29 | Slt Logic Llc | Programmable multi-service queue scheduler |
| US7151744B2 (en) * | 2001-09-21 | 2006-12-19 | Slt Logic Llc | Multi-service queuing method and apparatus that provides exhaustive arbitration, load balancing, and support for rapid port failover |
| US6980513B2 (en) * | 2001-09-24 | 2005-12-27 | Transwitch Corporation | Methods and apparatus for the fair allocation of bandwidth among MCR and best effort service connections in an ATM switch |
| EP1313274A3 (en) * | 2001-11-19 | 2003-09-03 | Matsushita Electric Industrial Co., Ltd. | Packet transmission apparatus and packet transmission processing method |
| US7352694B1 (en) * | 2001-12-14 | 2008-04-01 | Applied Micro Circuits Corporation | System and method for tolerating data link faults in a packet communications switch fabric |
| US7079545B1 (en) * | 2001-12-17 | 2006-07-18 | Applied Microcircuits Corporation ( Amcc) | System and method for simultaneous deficit round robin prioritization |
| US8418129B1 (en) | 2001-12-14 | 2013-04-09 | Qualcomm Incorporated | Method for automatically generating code to define a system of hardware elements |
| US7424013B1 (en) * | 2001-12-20 | 2008-09-09 | Applied Micro Circuits Corporation | System and method for granting arbitrated bids in the switching of information |
| JP2003188841A (en) * | 2001-12-17 | 2003-07-04 | Nec Microsystems Ltd | Digital data multiplexer |
| US7324452B2 (en) * | 2002-01-14 | 2008-01-29 | Fujitsu Limited | Weighted credit-based arbitration using credit history |
| US6807588B2 (en) * | 2002-02-27 | 2004-10-19 | International Business Machines Corporation | Method and apparatus for maintaining order in a queue by combining entry weights and queue weights |
| US7545814B2 (en) * | 2002-02-28 | 2009-06-09 | Nokia Corporation | Method and system for dynamic remapping of packets for a router |
| US7362749B2 (en) * | 2002-03-01 | 2008-04-22 | Verizon Business Global Llc | Queuing closed loop congestion mechanism |
| US20030200317A1 (en) * | 2002-04-19 | 2003-10-23 | Native Networks Technologies Ltd | Method and system for dynamically allocating bandwidth to a plurality of network elements |
| US7187687B1 (en) | 2002-05-06 | 2007-03-06 | Foundry Networks, Inc. | Pipeline method and system for switching packets |
| US7468975B1 (en) | 2002-05-06 | 2008-12-23 | Foundry Networks, Inc. | Flexible method for processing data packets in a network routing system for enhanced efficiency and monitoring capability |
| US20120155466A1 (en) | 2002-05-06 | 2012-06-21 | Ian Edward Davis | Method and apparatus for efficiently processing data packets in a computer network |
| US6901063B2 (en) * | 2002-05-13 | 2005-05-31 | Qualcomm, Incorporated | Data delivery in conjunction with a hybrid automatic retransmission mechanism in CDMA communication systems |
| US6822939B2 (en) | 2002-05-20 | 2004-11-23 | Transwitch Corporation | Method and apparatus for guaranteeing a minimum cell rate (MCR) for asynchronous transfer mode (ATM) traffic queues |
| FI115940B (en) * | 2002-06-10 | 2005-08-15 | First Hop Oy | Method and apparatus for implementing service quality in data transmissions |
| CN1314234C (en) * | 2002-07-05 | 2007-05-02 | 中兴通讯股份有限公司 | Multi-channel weiighted polling method and device |
| US7263065B1 (en) | 2002-08-30 | 2007-08-28 | At&T Corp. | Network having multiple QoS levels |
| US6901072B1 (en) | 2003-05-15 | 2005-05-31 | Foundry Networks, Inc. | System and method for high speed packet transmission implementing dual transmit and receive pipelines |
| US7551557B2 (en) * | 2003-06-19 | 2009-06-23 | International Business Machines Corporation | Method and system for ensuring arbitration fairness in bus ring topology |
| US20050059405A1 (en) * | 2003-09-17 | 2005-03-17 | Trapeze Networks, Inc. | Simulation driven wireless LAN planning |
| US20050059406A1 (en) * | 2003-09-17 | 2005-03-17 | Trapeze Networks, Inc. | Wireless LAN measurement feedback |
| US7701854B2 (en) * | 2003-11-17 | 2010-04-20 | International Business Machines Corporation | Differentiated handling of SIP messages for VoIP call control |
| JP2005204002A (en) * | 2004-01-15 | 2005-07-28 | Fujitsu Ltd | Packet relaying device |
| US7221927B2 (en) * | 2004-02-13 | 2007-05-22 | Trapeze Networks, Inc. | Station mobility between access points |
| US7817659B2 (en) | 2004-03-26 | 2010-10-19 | Foundry Networks, Llc | Method and apparatus for aggregating input data streams |
| US8730961B1 (en) | 2004-04-26 | 2014-05-20 | Foundry Networks, Llc | System and method for optimizing router lookup |
| US7929442B2 (en) * | 2004-06-30 | 2011-04-19 | Intel Corporation | Method, system, and program for managing congestion in a network controller |
| US9294353B2 (en) * | 2004-09-08 | 2016-03-22 | Cradlepoint, Inc. | Configuring a wireless router |
| US8732808B2 (en) * | 2004-09-08 | 2014-05-20 | Cradlepoint, Inc. | Data plan activation and modification |
| US20070254727A1 (en) * | 2004-09-08 | 2007-11-01 | Pat Sewall | Hotspot Power Regulation |
| US9584406B2 (en) * | 2004-09-08 | 2017-02-28 | Cradlepoint, Inc. | Data path switching |
| US20090172658A1 (en) * | 2004-09-08 | 2009-07-02 | Steven Wood | Application installation |
| US7764784B2 (en) * | 2004-09-08 | 2010-07-27 | Cradlepoint, Inc. | Handset cradle |
| US8477639B2 (en) | 2004-09-08 | 2013-07-02 | Cradlepoint, Inc. | Communicating network status |
| US9237102B2 (en) * | 2004-09-08 | 2016-01-12 | Cradlepoint, Inc. | Selecting a data path |
| US9232461B2 (en) * | 2004-09-08 | 2016-01-05 | Cradlepoint, Inc. | Hotspot communication limiter |
| FI118072B (en) | 2004-09-24 | 2007-06-15 | Tellabs Oy | A method and apparatus for scheduling transmission link capacity between packet switched communication networks maintaining retention characteristics |
| CN100421428C (en) * | 2004-10-28 | 2008-09-24 | 华为技术有限公司 | Method for scheduling forward direction public control channel message |
| US7657703B1 (en) * | 2004-10-29 | 2010-02-02 | Foundry Networks, Inc. | Double density content addressable memory (CAM) lookup scheme |
| US20060101178A1 (en) * | 2004-11-08 | 2006-05-11 | Zhong Tina C | Arbitration in a multi-protocol environment |
| US20060161733A1 (en) * | 2005-01-19 | 2006-07-20 | Emulex Design & Manufacturing Corporation | Host buffer queues |
| US7672303B1 (en) * | 2005-02-17 | 2010-03-02 | Emc Corporation | Arbitration method and system |
| US7529925B2 (en) * | 2005-03-15 | 2009-05-05 | Trapeze Networks, Inc. | System and method for distributing keys in a wireless network |
| US8638762B2 (en) | 2005-10-13 | 2014-01-28 | Trapeze Networks, Inc. | System and method for network integrity |
| US7551619B2 (en) | 2005-10-13 | 2009-06-23 | Trapeze Networks, Inc. | Identity-based networking |
| US7573859B2 (en) | 2005-10-13 | 2009-08-11 | Trapeze Networks, Inc. | System and method for remote monitoring in a wireless network |
| US7724703B2 (en) | 2005-10-13 | 2010-05-25 | Belden, Inc. | System and method for wireless network monitoring |
| WO2007044986A2 (en) | 2005-10-13 | 2007-04-19 | Trapeze Networks, Inc. | System and method for remote monitoring in a wireless network |
| US8250587B2 (en) | 2005-10-27 | 2012-08-21 | Trapeze Networks, Inc. | Non-persistent and persistent information setting method and system for inter-process communication |
| US20070130367A1 (en) * | 2005-11-18 | 2007-06-07 | Vivek Kashyap | Inbound connection prioritization |
| US8448162B2 (en) | 2005-12-28 | 2013-05-21 | Foundry Networks, Llc | Hitless software upgrades |
| US7609709B2 (en) * | 2006-01-23 | 2009-10-27 | Alcatel-Lucent Usa Inc. | Video aware traffic management |
| US7558266B2 (en) | 2006-05-03 | 2009-07-07 | Trapeze Networks, Inc. | System and method for restricting network access using forwarding databases |
| US20070260720A1 (en) * | 2006-05-03 | 2007-11-08 | Morain Gary E | Mobility domain |
| US8966018B2 (en) | 2006-05-19 | 2015-02-24 | Trapeze Networks, Inc. | Automated network device configuration and network deployment |
| US20070268515A1 (en) * | 2006-05-19 | 2007-11-22 | Yun Freund | System and method for automatic configuration of remote network switch and connected access point devices |
| US20070268506A1 (en) * | 2006-05-19 | 2007-11-22 | Paul Zeldin | Autonomous auto-configuring wireless network device |
| US20070268516A1 (en) * | 2006-05-19 | 2007-11-22 | Jamsheed Bugwadia | Automated policy-based network device configuration and network deployment |
| US20070268514A1 (en) * | 2006-05-19 | 2007-11-22 | Paul Zeldin | Method and business model for automated configuration and deployment of a wireless network in a facility without network administrator intervention |
| US7577453B2 (en) * | 2006-06-01 | 2009-08-18 | Trapeze Networks, Inc. | Wireless load balancing across bands |
| US7912982B2 (en) | 2006-06-09 | 2011-03-22 | Trapeze Networks, Inc. | Wireless routing selection system and method |
| US8818322B2 (en) | 2006-06-09 | 2014-08-26 | Trapeze Networks, Inc. | Untethered access point mesh system and method |
| US9258702B2 (en) | 2006-06-09 | 2016-02-09 | Trapeze Networks, Inc. | AP-local dynamic switching |
| US9191799B2 (en) | 2006-06-09 | 2015-11-17 | Juniper Networks, Inc. | Sharing data between wireless switches system and method |
| US7844298B2 (en) * | 2006-06-12 | 2010-11-30 | Belden Inc. | Tuned directional antennas |
| US7724704B2 (en) | 2006-07-17 | 2010-05-25 | Beiden Inc. | Wireless VLAN system and method |
| US8340110B2 (en) | 2006-09-15 | 2012-12-25 | Trapeze Networks, Inc. | Quality of service provisioning for wireless networks |
| US8072952B2 (en) | 2006-10-16 | 2011-12-06 | Juniper Networks, Inc. | Load balancing |
| US8054847B2 (en) * | 2006-10-31 | 2011-11-08 | Hewlett-Packard Development Company, L.P. | Buffer management in a network device |
| US20080107077A1 (en) * | 2006-11-03 | 2008-05-08 | James Murphy | Subnet mobility supporting wireless handoff |
| US8238255B2 (en) | 2006-11-22 | 2012-08-07 | Foundry Networks, Llc | Recovering from failures without impact on data traffic in a shared bus architecture |
| US7920592B2 (en) * | 2006-12-20 | 2011-04-05 | Lantiq Deutschland Gmbh | Method of bandwidth control and bandwidth control device |
| WO2008083339A2 (en) | 2006-12-28 | 2008-07-10 | Trapeze Networks, Inc. | Application-aware wireless network system and method |
| US7873061B2 (en) | 2006-12-28 | 2011-01-18 | Trapeze Networks, Inc. | System and method for aggregation and queuing in a wireless network |
| US7978614B2 (en) | 2007-01-11 | 2011-07-12 | Foundry Network, LLC | Techniques for detecting non-receipt of fault detection protocol packets |
| US9021081B2 (en) * | 2007-02-12 | 2015-04-28 | Cradlepoint, Inc. | System and method for collecting individualized network usage data in a personal hotspot wireless network |
| US8644272B2 (en) * | 2007-02-12 | 2014-02-04 | Cradlepoint, Inc. | Initiating router functions |
| US8037399B2 (en) | 2007-07-18 | 2011-10-11 | Foundry Networks, Llc | Techniques for segmented CRC design in high speed networks |
| US8271859B2 (en) | 2007-07-18 | 2012-09-18 | Foundry Networks Llc | Segmented CRC design in high speed networks |
| US8902904B2 (en) | 2007-09-07 | 2014-12-02 | Trapeze Networks, Inc. | Network assignment based on priority |
| US8509128B2 (en) | 2007-09-18 | 2013-08-13 | Trapeze Networks, Inc. | High level instruction convergence function |
| US8509236B2 (en) | 2007-09-26 | 2013-08-13 | Foundry Networks, Llc | Techniques for selecting paths and/or trunk ports for forwarding traffic flows |
| US8238942B2 (en) | 2007-11-21 | 2012-08-07 | Trapeze Networks, Inc. | Wireless station location detection |
| US8045563B2 (en) * | 2007-12-27 | 2011-10-25 | Cellco Partnership | Dynamically adjusted credit based round robin scheduler |
| US8335262B2 (en) * | 2008-01-16 | 2012-12-18 | Verivue, Inc. | Dynamic rate adjustment to splice compressed video streams |
| US8150357B2 (en) | 2008-03-28 | 2012-04-03 | Trapeze Networks, Inc. | Smoothing filter for irregular update intervals |
| US7885270B2 (en) * | 2008-05-15 | 2011-02-08 | Verlvue, Inc. | Statistical multiplexing of compressed video streams |
| US9106592B1 (en) * | 2008-05-18 | 2015-08-11 | Western Digital Technologies, Inc. | Controller and method for controlling a buffered data transfer device |
| US8474023B2 (en) | 2008-05-30 | 2013-06-25 | Juniper Networks, Inc. | Proactive credential caching |
| US7941578B2 (en) * | 2008-06-11 | 2011-05-10 | Hewlett-Packard Development Company, L.P. | Managing command request time-outs in QOS priority queues |
| US8978105B2 (en) | 2008-07-25 | 2015-03-10 | Trapeze Networks, Inc. | Affirming network relationships and resource access via related networks |
| US8238298B2 (en) | 2008-08-29 | 2012-08-07 | Trapeze Networks, Inc. | Picking an optimal channel for an access point in a wireless network |
| US8090901B2 (en) | 2009-05-14 | 2012-01-03 | Brocade Communications Systems, Inc. | TCAM management approach that minimize movements |
| US8325194B1 (en) | 2009-06-10 | 2012-12-04 | Nvidia Corporation | Mitigating main crossbar load using dedicated connections for certain traffic types |
| US8065465B1 (en) * | 2009-06-10 | 2011-11-22 | Nvidia Corporation | Mitigating main crossbar load using dedicated connections for certain traffic types |
| US9367487B1 (en) | 2009-06-10 | 2016-06-14 | Nvidia Corporation | Mitigating main crossbar load using dedicated connections for certain traffic types |
| US8264966B1 (en) * | 2009-09-04 | 2012-09-11 | Sprint Communications Company L.P. | Overload management on backhaul links based on packet loss on RF links |
| US8599850B2 (en) | 2009-09-21 | 2013-12-03 | Brocade Communications Systems, Inc. | Provisioning single or multistage networks using ethernet service instances (ESIs) |
| CN101784082A (en) * | 2009-12-22 | 2010-07-21 | 中兴通讯股份有限公司 | Method and device for enhancing service quality in wireless local area network |
| US20110242978A1 (en) * | 2010-03-31 | 2011-10-06 | Alcatel-Lucent Usa, Incorporated | System and method for dynamically adjusting quality of service configuration based on real-time traffic |
| US8937858B2 (en) * | 2010-09-27 | 2015-01-20 | Telefonaktiebolaget L M Ericsson (Publ) | Reducing access network congestion caused by oversubscription of multicast groups |
| US8542836B2 (en) | 2010-12-01 | 2013-09-24 | Juniper Networks, Inc. | System, apparatus and methods for highly scalable continuous roaming within a wireless network |
| JP2011147196A (en) * | 2011-05-06 | 2011-07-28 | Hitachi Ltd | Network node |
| CA2750345C (en) * | 2011-08-24 | 2013-06-18 | Guest Tek Interactive Entertainment Ltd. | Method of allocating bandwidth between zones according to user load and bandwidth management system thereof |
| CN102932275B (en) * | 2012-11-20 | 2015-04-22 | 中国空间技术研究院 | Priority message forwarding method applied to allowed time delay network |
| US9674086B2 (en) * | 2013-11-05 | 2017-06-06 | Cisco Technology, Inc. | Work conserving schedular based on ranking |
| CN104852866B (en) * | 2014-02-17 | 2019-04-30 | 深圳市中兴微电子技术有限公司 | A kind of output arbitration control method and device |
| US10069745B2 (en) * | 2016-09-12 | 2018-09-04 | Hewlett Packard Enterprise Development Lp | Lossy fabric transmitting device |
| US20200004584A1 (en) * | 2018-06-28 | 2020-01-02 | William Burroughs | Hardware Queue Manager for Scheduling Requests in a Processor |
| KR102165864B1 (en) * | 2019-07-22 | 2020-10-14 | 성균관대학교산학협력단 | Methods and apparatuses for packet scheduling for software defined networking in edge computing environment |
Family Cites Families (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4001702A (en) * | 1975-11-03 | 1977-01-04 | R F L Industries, Inc. | High speed fm and am demodulator |
| US4475192A (en) * | 1982-02-16 | 1984-10-02 | At&T Bell Laboratories | Data packet flow control scheme for switching networks |
| US4630261A (en) * | 1984-07-30 | 1986-12-16 | International Business Machines Corp. | Integrated buffer management and signaling technique |
| US4623996A (en) * | 1984-10-18 | 1986-11-18 | Mcmillen Robert J | Packet switched multiple queue NXM switch node and processing method |
| US4847754A (en) * | 1985-10-15 | 1989-07-11 | International Business Machines Corporation | Extended atomic operations |
| US4901348A (en) * | 1985-12-24 | 1990-02-13 | American Telephone And Telegraph Company | Data transmission security arrangement for a plurality of data stations sharing access to a communication network |
| JPH01177239A (en) * | 1988-01-06 | 1989-07-13 | Nec Corp | Packet concentrator and packet switching device |
| US4894824A (en) * | 1988-03-31 | 1990-01-16 | American Telephone And Telegraph Company, At&T Bell Laboratories | Control network for a rapid connection circuit switch |
| US4862454A (en) * | 1988-07-15 | 1989-08-29 | International Business Machines Corporation | Switching method for multistage interconnection networks with hot spot traffic |
| US4914650A (en) * | 1988-12-06 | 1990-04-03 | American Telephone And Telegraph Company | Bandwidth allocation and congestion control scheme for an integrated voice and data network |
| US5001702A (en) * | 1989-09-26 | 1991-03-19 | At&T Bell Laboratories | Packet switching network for multiple packet types |
-
1990
- 1990-07-11 US US07/551,712 patent/US5231633A/en not_active Expired - Lifetime
-
1991
- 1991-06-24 JP JP3511790A patent/JP2593120B2/en not_active Expired - Lifetime
- 1991-06-24 EP EP91912635A patent/EP0494284B1/en not_active Expired - Lifetime
- 1991-06-24 DE DE69130286T patent/DE69130286T2/en not_active Expired - Lifetime
- 1991-06-24 CA CA002065043A patent/CA2065043C/en not_active Expired - Lifetime
- 1991-06-24 AU AU81803/91A patent/AU629757B2/en not_active Ceased
- 1991-06-24 WO PCT/US1991/004444 patent/WO1992001345A1/en active IP Right Grant
Also Published As
| Publication number | Publication date |
|---|---|
| DE69130286T2 (en) | 1999-09-02 |
| EP0494284A1 (en) | 1992-07-15 |
| EP0494284B1 (en) | 1998-09-30 |
| CA2065043A1 (en) | 1992-01-12 |
| AU8180391A (en) | 1992-02-04 |
| EP0494284A4 (en) | 1992-11-19 |
| DE69130286D1 (en) | 1998-11-05 |
| JPH05502776A (en) | 1993-05-13 |
| JP2593120B2 (en) | 1997-03-26 |
| AU629757B2 (en) | 1992-10-08 |
| US5231633A (en) | 1993-07-27 |
| WO1992001345A1 (en) | 1992-01-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CA2065043C (en) | Method for prioritizing selectively discarding and multiplexing differing traffic type fast packets | |
| EP0533900B1 (en) | Low delay or low loss cell switch for atm | |
| US5287347A (en) | Arrangement for bounding jitter in a priority-based switching system | |
| US5268900A (en) | Device and method for implementing queueing disciplines at high speeds | |
| US4769811A (en) | Packet switching system arranged for congestion control | |
| US5278828A (en) | Method and system for managing queued cells | |
| US5150358A (en) | Serving constant bit rate traffic in a broadband data switch | |
| CA1286758C (en) | Packet switching system arranged for congestion control through bandwidth management | |
| EP0683945B1 (en) | Method for adaptive smoothing delay for packet voice applications | |
| EP0920235A2 (en) | Congestion management in a multi-port shared memory switch | |
| US6912224B1 (en) | Adaptive playout buffer and method for improved data communication | |
| JP2002543740A (en) | Method and apparatus for managing traffic in an ATM network | |
| Hluchyj et al. | Queueing disciplines for integrated fast packet networks | |
| KR100221324B1 (en) | Apparatus and method of dynamic priority queueing discipline using the per-session frame defined by the synchronus counter operation in atm networks | |
| KR100223055B1 (en) | Data output buffer control apparatus in atm switch | |
| GB2305576A (en) | ATM network having a plurality of input buffers for congestion avoidance | |
| KR100482687B1 (en) | Congestion Control Apparatus And Method For UBR Service In ATM Switch | |
| GB2344974A (en) | Fair packet discard in networks | |
| Dagiuklas et al. | Priority queueing disciplines in ATM switches carrying two layer video traffic | |
| CA2271669A1 (en) | Method and system for scheduling packets in a telecommunications network | |
| Liu | Tra c Shaping for Congestion Control in High Speed ATM Networks | |
| Gan et al. | Performance of an ATM network with multimedia traffic-a simulation study | |
| Khorsandi et al. | Statistical multiplexing of VBR MPEG sources under credit-based flow control | |
| KR19980017793A (en) | Jitter Guaranteed Method in Hierarchical Round-Robin Queue Service in ATM Networks |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| EEER | Examination request | ||
| MKEX | Expiry |