CA2288772A1 - Method and apparatus for precedence and preemption in atm connection admission control - Google Patents

Method and apparatus for precedence and preemption in atm connection admission control Download PDF

Info

Publication number
CA2288772A1
CA2288772A1 CA002288772A CA2288772A CA2288772A1 CA 2288772 A1 CA2288772 A1 CA 2288772A1 CA 002288772 A CA002288772 A CA 002288772A CA 2288772 A CA2288772 A CA 2288772A CA 2288772 A1 CA2288772 A1 CA 2288772A1
Authority
CA
Canada
Prior art keywords
call
preempted
precedence
vcc
active
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.)
Abandoned
Application number
CA002288772A
Other languages
French (fr)
Inventor
Scott Michael Poretsky
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Ahead Communications Systems Inc
Original Assignee
Individual
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Individual filed Critical Individual
Publication of CA2288772A1 publication Critical patent/CA2288772A1/en
Abandoned legal-status Critical Current

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L49/00Packet switching elements
    • H04L49/20Support for services
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L49/00Packet switching elements
    • H04L49/25Routing or path finding in a switch fabric
    • H04L49/253Routing or path finding in a switch fabric using establishment or release of connections between ports
    • H04L49/254Centralised controller, i.e. arbitration or scheduling
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L12/00Data switching networks
    • H04L12/54Store-and-forward switching systems 
    • H04L12/56Packet switching systems
    • H04L12/5601Transfer mode dependent, e.g. ATM
    • H04L2012/5629Admission control
    • H04L2012/563Signalling, e.g. protocols, reference model
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L12/00Data switching networks
    • H04L12/54Store-and-forward switching systems 
    • H04L12/56Packet switching systems
    • H04L12/5601Transfer mode dependent, e.g. ATM
    • H04L2012/5638Services, e.g. multimedia, GOS, QOS
    • H04L2012/5646Cell characteristics, e.g. loss, delay, jitter, sequence integrity
    • H04L2012/5651Priority, marking, classes

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

A informational element is provided which permits the user to assign a precedence level to a call. A precedence/preemption connection admission control (60) is also provided for processing a virtual circuit connection request having an assigned precedence level, and for preempting one or more VCCs when a VCC request having a relatively higher precedence level is received at an ATM switch. Preferably, a resource allocator (68) containing two databases, one listing all active virtual circuit connections and a second listing all preempted virtual circuits, is provided for storing the call parameters of preempted VCCs. According to a preferred embodiment, preempted VCCs may be reestablished. Reestablishment occurs according to various criteria. In addition, a second embodiment of a P/P CAC is provided in which active VCCs identified for preemption are buffered until the requested VCC is accepted by a downstream ATM switch.

Description

METHOD AND APPARATUS FOR PRECEDENCE AND PREEMPTION
IN ATM CONNECTION ADMISSION CONTROL
BACKGROUND OF THE INVENTION
1. Field of the Invention The present invention relates broadly to the field of telecommunications. More particularly, the present invention relates to the acceptance or rejection of a proposed connection through an asynchronous transfer mode (ATM) switch or node based on available bandwidth and the precedence assigned to the proposed connection.
2. State of the Art Perhaps the most awaited, and now fastest growing technology in the field of telecommunications in the 1990's is known as Asynchronous Transfer Mode (ATM) technology. ATM is providing a mechanism for removing performance limitations of local area networks (LANs) and wide area networks (WANs) and providing data transfers at a speed of on the order of gigabits/second. The variable length packets of LAN and WAN
data are being replaced with ATM cells which are relatively short, fixed length packets. Because ATM cells can carry voice, video and data across a single backbone network, the ATM
technology provides a unitary mechanism for high speed end-to-end telecommunications traffic.
Because the data contained in the ATM cells can be generated from either generally fixed rate communications, or bursty type communications, it will be appreciated that traffic accommodation mechanisms have been introduced in order to avoid situations where ATM switches or nodes are over-taxed, resulting in loss of cells. In particular, buffering and leaky-bucket usage regulating mechanisms are well known. In addition, it is known in the art that ATM switches and nodes will first determine whether they have the capacity to handle a proposed call before accepting the call. This is achieved through connection admission control (CAC).

An important part of CAC is the signaling between the user and the network for call establishment. The user network interface (UNI) is the interface between the user and the network. There are five possible states for the UNI (null, call initiated, outgoing call proceeding, active, and release) and four signalling messages (SETUP, CAhL PROCEEDING, CONNECT, and RELEASE requests). The CAC performs a network defined algorithm to determine whether the call request can be accepted while maintaining the guaranteed quality of service (QoS) of each currently existing connection. This is determined based upon the service class and traffic descriptors in information fields of the SETUP message. If a call request from the user is acceptable to the network, then a CALL PROCEEDING message is sent to the user. CONNECT messages are then exchanged and the connection becomes active for data cells to flow on the newly established connection from the user to the network. If the network cannot meet the QoS of the existing connections with the addition of the new connection, then the call request is rejected and a RELEASE message is returned to the user.
Referring to prior art Figure 1(a), all of the call control signaling messages are provided in Q.2931 signaling format, with bytes 1 through 9 being standard, and bytes 10 and higher being message dependent information elements (IE). The signaling format for an IE is shown in Figure 1(b). The SETUP message includes a number of IEs which are mandatory (user cell rate, called party number, connection identifier, QoS parameters, and broadband bearer capability), and may include optional IEs.
Current optional IEs for the SETUP message include AAL
parameter, calling party number, end point reference, broadband higher layer information, and broadband lower layer information.
The bytes of the mandatory messages and optional IEs permit the CAC of an ATM switch to make an accept or reject decision.
Prior art Figure 2 shows the architecture of an ATM switch 10. Input modules 12, 14 extract the ATM cell stream, perform usage parameter control, check for cell errors, and pass the acceptable cells to the switch fabric 16. The switch fabric 15 switches the cells from the input modules 12, 14 to the proper output modules 18, 20 based upon their VPI/VCI value. Signaling cells are identified by their VPI/VCI value and switched from the switch fabric 16 to the CAC 22 for processing. The output modules 18, 20 perform the opposite function of the input modules in order to prepare the switched cells into streams of cells for transmission.
Once a call is accepted, ATM requires that the QoS agreed to in the traffic contract between the source and network be guaranteed. To this end, ATM employs preventive and reactive traffic control methods through CAC, performed in a separate module in an ATM switch, and usage parameter control (UPC), performed within the input modules to prevent calls from being accepted if the QoS cannot be guaranteed. CAC generally performs the following functions: (1) negotiating new connection requests with a user and establishing a traffic contract characterizing source traffic and QoS, (2) deciding on admission or rejection of the new connection according to the network policy, (3) allocating network resources so that the network efficiency is maximized with the addition of a new connection, (4) providing acceptable values to the UPC, and (5) releasing network resources when a virtual circuit is disconnected.
Referring to prior art Figure 3, the CAC block receives a SETUP message at 30, and based upon the mandatory and optional IEs, reads the necessary resources for a requested virtual circuit connection (VCC) and runs a bandwidth allocation algorithm at 32 to determine at 34 whether the necessary resources are available at the switch. If the bandwidth . allocation algorithm determines that the required resources are available, the CAC updates at 36 the allocation database with the new VCC and allocated resources, a traffic contract is agreed to at 38, and the VCC is passed at 40 to the user in a CONNECT message. If the algorithm determines that the required resources are not available, the CAC rejects the call at 42.
The rejection is communicated through a RELEASE message. A
number of different bandwidth allocation algorithms have been used for CAC, with the goal of each algorithm being to maximize the admission region and statistical gain of one or more particular types of ATM traffic over a switch without exceeding the bandwidth of the switch.
In addition, several algorithms have been proposed in which the ATM network assigns one of two precedence levels to cells traversing a VCC. For example, in 1992, Nippon Telephone and Telegraph proposed a buffer reservation scheme in which the cells of a call are assigned to one of two levels of precedence (Saito, H., "Hybrid Connection Admission Control in ATM
Networks", SuperComm/ICC, 1992). A precedence would be assigned to a call using the cell loss priority (CLP) bit. The amount of buffer space reserved for high precedence cells is dynamically adjusted according to the required bandwidth of the high precedence cells. Low precedence cells would have available the remaining buffer space, in the dynamically allocated buffer.
However, it is only once the network has determined that it can accept a requested VCC, that precedence is given to the high precedence cell stream.
Bellcore and Brooklyn Polytechnic Institute in 1992 proposed that service classes be treated as precedence levels and that a cell scheduling policy be implemented at the output buffer of the ATM switch to ensure that the traffic descriptor of the precedence levels are met (Chao, J. and Uzun, N., "An ATM
Queue Manager with Multiple Delays and Loss Priorities", IEEE
Globecom, 1992). Toshiba has proposed the same concept (see Esaki, H., "Call Admission Control Method for ATM Networks", SuperComm/ICC, 1992). However, while providing precedence for certain cells is helpful for guaranteeing QoS for certain service classes, these schemes do not guarantee that lower precedence calls will not cause a higher precedence user to fail to gain access to the network.
In 1995 AT&T Bell Labs proposed multiplexing output buffers and assigning each output buffer a precedence level (Elwalid, A.

and Mitra, D., "Analysis, Approximations, and Admission Control of a Multi-Service Multiplexing System with Priorities", IEEE, 1995). Buffer access to the switch output would be determined by the precedence level of the buffer and the status of higher precedence buffers. This approach has several disadvantages.
First, the buffer sizes are pre-allocated. Second, precedence is based solely upon cell traffic already on the network, as no consideration is made of the precedence levels of the cell traffic attempting to be connected to the network which may be higher than other cell traffic already on the network.
Columbia University and ChipCom have proposed dedicating service buffers coupled with dynamic precedence-based allocation within each buffer (Dailianas, A. and Bovopoulos, A., "Design of a Real-Time Call Admission Control Algorithm with Priority Support", IEEE, 1995). As in the Nippon proposal, two levels of precedence are indicated by the CLP bit. The network measures the occurrences of high precedence cells based on their CLP bit value and dynamically adjusts the amount of space dedicated to high precedence cells within each service buffer. However, like each of the other proposed schemes, precedence is only assigned to cells on accepted traffic, and is not assigned to VCCs.
Preemption of a low precedence VCC for a higher precedence VCC
is not provisioned.
SUMMARY OF THE INVENTION
It is therefore an object of the invention to provide a method and apparatus for permitting multiple precedence levels to be assigned to virtual circuit connections over an ATM
switch.
It is another object of the invention to provide a method and apparatus for permitting precedence levels to be assigned by a user of the ATM network.

It is also an object of the invention to provide a method and apparatus for assigning more than two precedence levels to virtual circuit connections.
It is further object of the invention to provide a method and apparatus for preempting an existing virtual circuit having a certain precedence level in place of a call request having a user-assigned relatively higher precedence level.
It is yet another object of the invention to provide a method and apparatus for storing preempted calls in a database and automatically reestablishing the preempted calls in order of their precedence level as switch resources become available.
It is still a further object of the invention to provide a method and apparatus for buffering preempted calls slated to be preempted, and releasing preempted calls only when a complete virtual channel is established.
In accord with these objects which will be discussed in detail below, an new optional informational element (IE) is provided which permits the user to assign a precedence level to a VCC request. The IE is provided with a new IE identifier coding the IE for precedence levels. In a preferred embodiment, the network permits up to 256 user-assignable precedence levels to be assigned to a VCC, the precedence level of the VCC being coded by an octet of the contents of the IE.
In addition, a precedence/preemption connection admission control (P/P CAC) algorithm is provided for processing the precedence level assigned to a requested VCC and for preempting one or more VCCs when a requested VCC having a relatively higher precedence level is received at an ATM switch. The provided P/P
CAC algorithm is compatible with all bandwidth allocation algorithms of the art. In implementing a preferred embodiment of the P/P CAC at a switch, a resource allocator containing two databases is provided, one listing the parameters of all active VCCs and a second listing the parameters of all preempted VCCs.

According to a preferred embodiment, preempted VCCs may be reestablished according to different criteria.
Additional objects and advantages of the invention will become apparent to those skilled in the art upon reference to the detailed description taken in conjunction with the provided figures.
BRIEF DESCRIPTION OF THE DRAWINGS
Figures 1(a) and 1(b) show the prior art Q.2931 signaling format of a message and information element, respectively;
Figure 2 is a functional block diagram of prior art ATM
switch architecture;
Figure 3 is a typical connection admission control flow chart;
Figure 4 is a exemplary information element for coding a precedence level;
Figure 5 is a functional block diagram of the precedence and preemption connection admission control system according to the invention;
Figure 6 is a flow diagram of a precedence/preemption connection admission control system having a preempted VCC
database according to the invention according to a first embodiment of the invention;
Figure 7 is a flow diagram of a precedence/preemption connection admission control system having delayed preemption according to a second embodiment of the invention; and Figure 8 is a flow diagram of a precedence/preemption admission control system which attempts to reestablish preempted calls when resources are available according to a third embodiment of the invention.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
Turning to Figure 4, an informational element (IE) is shown in Q.2931 signalling format. The IE is provided for permitting a user to assign a precedence level to a call. The IE is provided with a particular IE identifier which codes the element as one containing a user assigned precedence level. In a preferred embodiment, one octet of the contents of the IE is provided for permitting the user to assign any of 256 precedence levels to a call. If a greater number of precedence levels are desired, more than one octet of contents may be used for precedence level assignment. Conversely, one or more bits of an octet may be used if the network requires fewer than 256 precedence levels.
Referring now to Figure 5 and according to the invention, a precedence/preemption connection admission control system (P/P
CAC) 60 is provided for processing the precedence level of a requested VCC and for preempting one or more active VCCs when a VCC request having a relatively higher precedence level is received at an ATM switch. The P/P CAC 60 includes four components: a message processor 62, a bandwidth allocation controller 64, a CAC algorithm processor 66, and a resource allocator 68. The message processor (MP) 62 receives Q.2931 protocol signals and parses received SETUP messages for requested QoS parameters, receives RELEASE messages when VCCs on the switch are terminated, and further initiates RELEASE
messages for rejected VCC requests and preempted VCCs. The message processor negotiates with a VCC requestor 61. The bandwidth allocation controller (BAC) 69 maintains the bandwidth assigned to each precedence level, receives VCC requests, forwards the required parameters to the bandwidth allocation algorithm processor (BAAP) 66, requests preemption and bandwidth reallocation to the resource allocator, and invokes a network initiated RELEASE message to a preempted user. The BAAP 66 selects the appropriate bandwidth allocation algorithm from among those available in its memory and runs the algorithm to make a VCC accept or VCC reject decision. The P/P CAC 60 operates with any bandwidth allocation algorithms) supplied by the network provider. The resource allocator (RA) 68 contains a first database 70 which lists all active virtual circuit connections (active VCC) and, according to a preferred embodiment of the invention, includes a second database 72 which lists all preempted virtual circuits connections (preempted VCC). The first and second databases 70, 72 may be provided in a single memory means or in two separate memory means.
Referring to Figures 5 and 6, and according to a first embodiment of the invention, a precedence and preemption connection admission control (P/P CAC) algorithm is provided for processing incoming VCC requests. In particular, the first embodiment is designed for activation upon receipt at the message processor of a SETUP message. First, the message processor 62 receives at 100 a SETUP message for a requested VCC
from the requestor 61. The message processor 62 analyzes the SETUP message, and determines the parameters of the VCC, including the precedence level of the VCC. The message processor sends the precedence level to the resource allocator 68, where the preempted VCC database 72 is examined at 102 for higher precedence VCCs that were previously preempted.
If parameters for a higher precedence level VCC are identified at 104 in the preempted VCC database 72, a RELEASE
message is sent at 106 from the message processor 62 back to the VCC requestor, and the requested VCC parameters are replaced at 108 with the parameters of the higher precedence level VCC in the preempted VCC database 72 to be considered later for re-admission. It will be appreciated that, under this first embodiment, consideration for re-admitting a previously preempted VCC occurs on the arrival of a new VCC request.

Therefore, the interval for considering a preempted VCC for re-admission is non-deterministic. However, VCC requests have been shown to arrive with a Poisson distribution. As a result, preempted VCCs will also be reconsidered for admission according to a Poisson distribution.
If a higher precedence VCC is not identified at 104 in the preempted VCC database 72, or after requested VCC parameters are replaced at 108 with the parameters of the higher precedence level VCC in the preempted VCC database 72, an updatable counter is set at 110 to zero. The counter represents the number of VCCs which will be preempted if the current VCC request is accepted. The bandwidth allocation controller 64 next activates the BAAP 66 to perform at 112 a bandwidth allocation algorithm.
If the first time through the P/P CAC algorithm (i.e., the counter at 116 remains set to zero) the bandwidth allocation algorithm designates the call for acceptance at 114 without preempting an existing VCC, the VCC is accepted, the active VCC
database 70 is updated with the new call parameters and a determination is made at 117 as to whether the VCC request came from a SETUP message or from the preempted VCC database. If the request came from a SETUP message, a CONNECT message is returned to the requestor and a SETUP message is forwarded to the next switch at 118. If the request came from the preempted VCC
database, the network manager sends a SETUP message at 119 to both the original source and destination.
If the bandwidth allocation algorithm rejects at 114 the call request, the counter is incremented at 120 by one. The counter is then compared at 122 to a predetermined (set by the network manager) maximum number of VCC preemptions. If the counter exceeds at 122 the maximum number of VCC preemptions, link parameters are restored at 122. Then, if VCC request is, at 123, for the VCC requested by the SETUP message (at 100), a RELEASE message is returned at 124 to the requestor 61.
Otherwise, the request is from the preempted VCC database, and the parameters for the VCC are returned at 125 to the preempted VCC database. If the counter does not exceed the maximum number of preemptions at 122, then a search is made at 126 in the active VCC database 70 for the lowest precedence VCC, and in the case of a tie, the VCC being most recently established. If no such call can be identified at 128 in the active VCC database, i.e., there are no active calls having a lower precedence than the requested call, then either a RELEASE message is returned at 124 (if the call request originated from a SETUP message) or the call parameters are returned at 125 to the preempted VCC
database (if the call request originated from the preempted VCC
database). If a call having lower precedence can be identified at 128, then the characteristics of the identified VCC are "subtracted" (removed) at 130 from the dynamic link parameters, and the P/P CAC algorithm repeats steps 112 onward. That is, the P/P CAC algorithm will continue to cycle through steps 112 through 130 in an attempt to identify link resources to be freed from lower precedence VCCs either until sufficient resources are available from active lower precedence VCCs or until the maximum number of VCCs permitted to be preempted has been exceeded. It is noted that each time through, the bandwidth allocation algorithm is conducted minus an additional VCC identified for preemption.
If during the above-discussed cycling, the bandwidth allocation algorithm accepts at 114 a VCC request prior to the counter exceeding the maximum number of preempted VCCs, then the VCC or VCCs identified at 128 for preemption are removed at 132 from the active VCC database 72 and placed into the preempted VCC database 70, and one or more RELEASE messages are sent at 132 to the VCC(s) identified. If the VCC request originated from a SETUP message, a VCC CONNECT message is provided to the requestor at 118 and a SETUP message is also forwarded at 118 to the next switch or node. On the other hand, if the VCC request originated from the preempted VCC database, a SETUP message is sent at 119 by the network to the original source (original requestor) and destination to reestablish the connection.
As the VCCs identified for preemption are stored in the preempted VCC database 70, the preempted VCCs will be considered at 102 for re-admission upon receipt of the next SETUP message for a VCC at the switch or node.
It will be appreciated that the P/P CAC algorithm considers user precedence levels, effective bandwidth, and requested service parameters for the VCC accept/reject decision. Higher precedence VCCs are permitted to preempt up to a predetermined number of lower precedence VCCs when resources are immediately unavailable for the higher precedence VCC, and in accord with the preferred embodiment of the invention, VCCs set for preemption are selected by a last in, first out rule. It will be appreciated that the process time for the algorithm is deterministic and any bandwidth allocation algorithm may be used with the P/P CAC.
It will be appreciated that it is not practical or even preferable to permanently store calls preempted in the preempted VCC database. Preempted calls will become "stale" after a certain period of time, i.e., the disconnected parties will no longer desire re-connection. Therefore, it is preferable that preempted calls be stored for only a certain period of time, e.g., one hour. Furthermore, while preferred, it is not required that the P/P CAC store preempted VCCs in a preempted VCC database. A modified and simpler algorithm may be provided omitting steps 102 through 108; i.e., after a SETUP message is received at 100, the counter is set at 110 to zero, and the algorithm continues forward therefrom. The algorithm is simpler and quicker to implement, and requires fewer resources at a switch or node, e.g., no memory for a preempted VCC database is required. However, omitting the storage of preempted VCCs in a preempted VCC database prevents the possibility of having the network reestablish a preempted VCC.
Turning to Figure 7, a second embodiment of a P/P CAC is provided. The P/P CAC of Figure 7 is substantially similar to the first embodiment (with like systems and steps having similar numbers incremented by 100), but further includes an option to delay preemption of VCCs until the requested VCC has been accepted at its destination. Upon receiving a SETUP message at 200, the P/P CAC checks at 202 the preempted VCC database for a previously preempted higher precedence VCC. If a higher precedence VCC is identified at 204, the requested VCC is provided at 206 with a RELEASE message and the identified VCC is replaced at 208 with the requested VCC. Depending on whether a higher precedence VCC is identified, requested VCC parameters are either the originally requested VCC or the replaced VCC
parameters. The counter is set at 210 to zero and a bandwidth allocation algorithm is run at 212 a first time. If the bandwidth allocation algorithm determines at 214 that sufficient resources are available to accept the requested VCC, a determination is then made at 217 as to whether the call request originated from a SETUP message or from the preempted VCC
database. If the request originated from a SETUP message, a CONNECT message is sent at 218 and the SETUP message is forwarded at 218 to the next switch or node. If the request originated from the preempted VCC database, a SETUP message is sent at 219 to the original source and destination. In either case, no VCCs are preempted. If the bandwidth allocation algorithm rejects at 214 the VCC, the counter is incremented at 220. While the counter at 221 does not exceed a maximum number of VCCs for preemption, the most recent lower precedence calls, if available, are identified at 226, 228 for preemption, and the parameters of the identified VCCs are subtracted at 230 from the occupied link resources. If the counter has exceeded at 221 the maximum number of VCCs for preemption and still the required resources are not available, the link parameters are restored at 222 and either a RELEASE message is returned at 224 (if the call request originated from a SETUP message) or the call parameters are returned at 225 to the preempted VCC database (if the call request originated from the preempted VCC database). Likewise, according to the above criteria at 223, if a VCC having a lower precedence cannot be identified at 228, the link parameters are restored at 222, and either a RELEASE message is sent at 224, or the parameters for the preempted VCC are returned to the preempted VCC database at 225.

If the bandwidth allocation algorithm is able to accept a requested VCC at 214 such that preemption of active VCCs is required (i.e., the counter has been incremented at least once to a value greater than zero), a SETUP message is forwarded at 234 to the downstream switch and the P/P CAC awaits at 236 a message from the downstream switch. If no message is returned at 236 before a "timeout", or if a RELEASE message is returned at 238 from the downstream switch or node prior to the "timeout", the link parameters are restored at 222 and either a RELEASE message is returned at 229, or the VCC parameters are returned at 225 to the preempted VCC database, depending on whether, at 223, the VCC request originated from a SETUP message or from the preempted VCC database. On the other hand, if a CONNECT message is received at 238 prior to the "timeout" from the downstream switch or node, the one or more VCCs identified for preemption are preempted in the active VCC database and transferred to the preempted VCC database at 240, and RELEASE
messages are sent, also at 290, for the preempted calls. A
CONNECT message is then sent at 244 for the VCC if the request originated from a SETUP message, a determination of which is made at 242. If the request originated from the preempted VCC
database, also determined at 242, a SETUP message is sent at 296 to the original source and destination.
In the above P/P CAC, the CONNECT message is not transmitted once the VCC is accepted on the basis of preemption.
Rather, the switch holds the CONNECT message until it receives a CONNECT from the downstream switch. Likewise, the RELEASE
message for the preempted VCC(s) is held until the CONNECT is received. As a result, a requested VCC is prevented from preempting active VCCs if the requested VCC is rejected at another switch in the network.
As stated above with respect to the first embodiment, it is not required that the P/P CAC store preempted VCCs in a preempted VCC database. Omitting the storage of preempted VCC
in a preempted VCC database will, as stated above, prevent the possibility of having the network reestablish a preempted call.

However, the algorithm is simpler and quicker to implement, and requires fewer resources.
Turning to Figure 8, a third embodiment of a precedence/
preemption connection admission control system is provided.
Unlike the first and second embodiments, which attempt to reestablish preempted VCCs upon receipt of a VCC SETUP request, the third embodiment uses an algorithm which attempts to reestablish preempted VCCs whenever a VCC is released from the switch; i.e., activation of the algorithm occurs upon receipt of a RELEASE message at the message processor. It will be appreciated from the outset that the third embodiment may be used in conjunction with either of the first two embodiments to provide an optimally functioning P/P CAC.
When the message processor receives at 300 a RELEASE
message, the released VCC is removed from the active VCC
database and the dynamic link parameters of resources utilized at the switch are updated at 301. A search is then made at 302 of the preempted VCC database for preempted VCCs. If a preempted VCC is not located in the preempted VCC database, the algorithm terminates at 304, awaiting the next RELEASE message.
If a preempted VCC is located at 302 in the preempted VCC
database, the preempted VCC having the highest precedence, and, in the case of a tie, which has been in the preempted VCC
database the shortest time, has its QoS parameters provided at 306 to the bandwidth allocation algorithm 308. The bandwidth allocation algorithm determines at 308 whether resources are available on the link for the previously preempted VCC and either accepts or rejects at 310 the VCC based upon the available resources. If the necessary resources are available at 310, the VCC is accepted at 312 for reestablishment, a SETUP
message initiated by the network is sent to the original source and destination, and the VCC is removed from the preempted VCC
database and added to the active VCC database. Another search is then made at 314 for other preempted VCCs in the preempted VCC database. If other preempted VCCs are located, the VCC
highest in precedence, and, in the case of a tie, having been in the preempted VCC shortest is chosen at 316 and sent to the bandwidth allocation algorithm 308 and continue to be processed according to P/P CAC algorithm (310-316). If a VCC request is rejected at 310 by the bandwidth allocation algorithm, another search is made at 314 to determine whether other preempted VCCs are in the preempted VCC database. Preferably, the entire preempted VCC database will thereby be searched, in order of precedence, to determine whether any of the preempted VCC may be reestablished. Once the entire preempted VCC database is searched, the algorithm terminates at 304, to be initiated again once the message processor receives the next RELEASE message.
Alternatively, only a portion of the preempted VCC database may be searched. For example, the P/P CAC may be set such that only preempted VCCs within two precedence levels of the highest precedence VCC in the preempted VCC database are searched.
When the P/P CAC of the third embodiment is used in conjunction with the P/P CAC of the first and second embodiments, it is preferable that the P/P CAC of the third embodiment be automatically interrupted when a SETUP message is received at the message processor. This is because SETUP
messages are preferably given priority over preempted VCCs, at least until their respective precedence levels are compared.
Using the algorithm of the third embodiment in conjunction with the algorithm of either the first or second embodiments, a P/P CAC system is provided wherein requested VCCs are given access to link resources based upon the precedence level of the requested VCCs relative to the precedence level of active VCCs currently traversing an ATM switch. Lower precedence level VCC
are preempted in favor of higher precedence VCCs. VCCs which have been preempted are attempted to be reestablished when bandwidth resources become available (i.e., upon receipt of RELEASE messages at the message processor) and prior to processing a requested VCC having a relatively lower precedence (i.e., upon receipt of a SETUP message at the message processor) .

There have been described and illustrated herein several embodiments of a precedence and preemption connection admission control. While particular embodiments of the invention have been described, it is not intended that the invention be limited thereto, as it is intended that the invention be as broad in scope as the art will allow and that the specification be read likewise. Thus, while an information element capable of containing any of 256 precedence levels has been disclosed, it will be appreciated that 256 levels need not be available; i.e., fewer than all 256 levels may be assignable. For example, only five distinct precedence levels may be desired. Moreover, additional octets of the informational element may be utilized to provide a greater number of precedence levels. Also, the preempted VCC database need not store all preempted VCCs, but may store only a certain number of preempted VCCs. Likewise, the active VCC database need not store all active VCCs.
However, it preferable that at least the VCCs having a precedence level other than the highest precedence level be stored in the active VCC database. It will also be appreciated that the third embodiment of the precedence and preemption connection admission control system may be modified to permit preemption of active calls by the previously preempted calls seeking reestablishing, as a similar process is described with respect to the first and second embodiments. In addition, further modifications to the precedence and preemption connection admission control can be made with reference to "An Algorithm for Precedence and Preemption in Asynchronous Transfer Mode (ATM) Call Admission Control" by Poretsky, et al. which is published by IEEE Infocomm and which is hereby incorporated by reference herein in its entirety. It will therefore be appreciated by those skilled in the art that yet other modifications could be made to the provided invention without deviating from its spirit and scope as so claimed.

Claims (22)

We Claim:
1. A method of assigning a precedence level to a call over an ATM switch, comprising:
a) providing a call message with an information element identifier indicating said information element includes precedence level information; and b) assigning a precedence level to said call message by coding at least one bit at a predetermined location in said information element.
2. A method according to claim 1, wherein:
said call message is an ATM SETUP message.
3. A method according to claim 1, wherein:
said at least one bit is a plurality of bits.
4. A method of preempting active calls traversing a switch or node in favor of an inactive call, comprising:
a) for each active call, storing a first indication of a precedence level assigned to that active call such that a plurality of first indications are stored;
b) determining a second indication of a precedence level assigned to the inactive call;
c) comparing said second indication with at least one of said plurality of first indications; and d) when said comparing indicates that said second indication indicates a relatively higher precedence than at least one of said plurality of first indications, designating said inactive call for acceptance, and designating at least one of said active calls having a first indication of relatively lower precedence than said second indication for preemption.
5. A method according to claim 4, further comprising:
e) accepting said inactive call designated for acceptance and preempting said active call designated for preemption when preempting provides sufficient resources at the switch or node for accepting said inactive call.
6. A method according to claim 5, further comprising:
f) using a bandwidth allocation algorithm to determine whether sufficient resources are available, said bandwidth allocation algorithm being performed on said active calls plus the inactive call designated for acceptance minus said active call designated for preemption.
7. A method according to claim 4, wherein:
said inactive call is a call requesting acceptance to traverse the switch or node.
8. A method according to claim 4, further comprising:
e) providing said preempted calls to a preempted call database.
9. A method according to claim 8, wherein:
said inactive call is a previously preempted call whose second indication of precedence level is located in said preempted call database.
10. A method according to claim 8, further comprising:
f) attempting to reestablish one of said preempted calls and rejecting an inactive call seeking acceptance when said one of said preempted calls in said preempted call database has a relatively higher precedence than said inactive call seeking acceptance.
11. A method according to claim 4, wherein:
said inactive call designated for acceptance is refused acceptance if accepting said inactive call designated for acceptance requires preempting more than a predetermined number of active calls.
12. A method according to claim 5, further comprising:
f) attempting reestablishment of said preempted call whenever resources are freed from the switch or node.
13. A method according to claim 12, wherein:
an inactive call designated for acceptance does not preempt an active call unless a destination for said inactive call accepts said call designated for acceptance.
14. A method according to claim 12, further comprising:
e) postponing reestablishment of said preempted call unless a destination accepts said preempted call prior to a timeout.
15. A method according to claim 5, further comprising:
f) attempting reestablishment of said preempted call whenever an inactive call seeking acceptance is received at the switch or node.
16. A method according to claim 15, wherein:
an inactive call designated for acceptance does not preempt an active call unless a destination for said inactive call accepts said call designated for acceptance.
17. A method according to claim 15, further comprising:
e) postponing reestablishment of said preempted call unless a destination accepts said preempted call prior to a timeout.
18. A method according to claim 4, wherein:
said active calls are ATM calls over an ATM switch or ATM
node.
19. A precedence and preemption connection admission control system for processing a call request in light of available resources at a switch or node, the call request having a precedence level, said system comprising:
a) a memory means including a first database means for storing information on active calls; and b) a processor means coupled to said memory means, said processor means i) for receiving the call request and reading parameters of the call request, ii) for determining whether sufficient resources are present at the switch or node for accepting the call request based on said information on active calls and based upon said parameters of said call request, iii) for preempting at least one active call having a relatively lower precedence level than the precedence level of the call request in order to accept the call request, and iv) for updating said information in said first database to add information on the call request and to delete information on said at least one active call preempted.
20. A system according to claim 19, wherein:
said memory means includes a second database means for storing second information on preempted calls, and said processor means includes means for updating said second database to add information on said at least one active call preempted.
21. A system according to claim 20, further comprising:
c) means for reestablishing said at least one active call preempted using information from said second database.
22. A system according to claim 19, wherein:
said processor means is adapted to receive an ATM call request, and said precedence level is provided in an information element of an ATM SETUP message.
CA002288772A 1997-05-09 1998-05-07 Method and apparatus for precedence and preemption in atm connection admission control Abandoned CA2288772A1 (en)

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
US08/853,654 1997-05-09
US08/853,654 US6141322A (en) 1997-05-09 1997-05-09 Method and apparatus for precedence and preemption in ATM connection admission control
PCT/US1998/009399 WO1998051060A1 (en) 1997-05-09 1998-05-07 Method and apparatus for precedence and preemption in atm connection admission control

Publications (1)

Publication Number Publication Date
CA2288772A1 true CA2288772A1 (en) 1998-11-12

Family

ID=25316586

Family Applications (1)

Application Number Title Priority Date Filing Date
CA002288772A Abandoned CA2288772A1 (en) 1997-05-09 1998-05-07 Method and apparatus for precedence and preemption in atm connection admission control

Country Status (3)

Country Link
US (1) US6141322A (en)
CA (1) CA2288772A1 (en)
WO (1) WO1998051060A1 (en)

Families Citing this family (35)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6968379B2 (en) * 1997-05-30 2005-11-22 Sun Microsystems, Inc. Latency-reducing bandwidth-prioritization for network servers and clients
US6097726A (en) * 1997-09-18 2000-08-01 Ascend Communications, Inc. Virtual path merging in a multipoint-to-point network tunneling protocol
US6157614A (en) * 1997-10-22 2000-12-05 Netro Corporation Wireless ATM network with high quality of service scheduling
US6345038B1 (en) * 1998-05-12 2002-02-05 International Business Machines Corporation Improving access to congested networks
FI105968B (en) * 1998-05-28 2000-10-31 Nokia Networks Oy Internal traffic in the communication node
AU8536198A (en) * 1998-06-05 1999-12-30 Nokia Networks Oy Method and apparatus for performing a connection admission control
US6633539B1 (en) * 1998-08-28 2003-10-14 Cisco Technology, Inc. Device, method and article of manufacture for call setup pacing in connection-oriented networks
JP3090131B2 (en) * 1998-11-20 2000-09-18 日本電気株式会社 Connection hold control method and control circuit in ATM exchange
US6771648B1 (en) * 1999-03-31 2004-08-03 Marconi Communications, Inc. Method and apparatus for a dynamic call admission control algorithm
US6711129B1 (en) * 1999-10-26 2004-03-23 Avaya Technology Corp. Real-time admission control
KR100651373B1 (en) 1999-11-24 2006-11-28 삼성전자주식회사 A method for avoiding over-flooding due to the address registration from unstable terminals
WO2001054354A1 (en) * 2000-01-20 2001-07-26 Mci Worldcom, Inc. Intelligent network and method for providing voice telephony over atm and point-to-multipoint connectivity
US6865150B1 (en) 2000-04-06 2005-03-08 Cisco Technology, Inc. System and method for controlling admission of voice communications in a packet network
US7277384B1 (en) 2000-04-06 2007-10-02 Cisco Technology, Inc. Program and method for preventing overload in a packet telephony gateway
DE60040329D1 (en) * 2000-05-09 2008-11-06 Lucent Technologies Inc Improved service quality control in a mobile telecommunications network
DE10063390A1 (en) * 2000-12-19 2002-06-20 Bosch Gmbh Robert Method for controlling connections in a communication system therefor
US7058031B2 (en) * 2001-01-31 2006-06-06 Qualcomm Incorporated Method and apparatus for efficient use of communication resources in a data communication system under overload conditions
US7257125B1 (en) 2002-01-22 2007-08-14 Marvell International Ltd. Quality of service half-duplex media access controller
JP2003229877A (en) * 2002-01-31 2003-08-15 Fujitsu Ltd Net side transmission equipment and method for controlling variable rate traffic by allocation of dynamic band
US8239534B1 (en) * 2003-07-14 2012-08-07 Lockheed Martin Corporation Precedence adjusted resource allocation
US8554860B1 (en) * 2003-09-05 2013-10-08 Sprint Communications Company L.P. Traffic segmentation
US8750289B2 (en) 2003-10-01 2014-06-10 Itxc Ip Holdings S.A.R.L. Call setup using voice over the internet protocol (VoIP)
JP4053971B2 (en) * 2003-11-28 2008-02-27 株式会社東芝 Telephone exchange device and method for controlling telephone exchange device
US7308617B2 (en) * 2004-06-17 2007-12-11 International Business Machines Corporation Apparatus, system, and method for automatically freeing a server resource locked awaiting a failed acknowledgement from a client
US7443792B2 (en) * 2004-08-27 2008-10-28 Cisco Technology, Inc. Method and system for selecting connections to bump based on priority in a network
FR2875667A1 (en) * 2004-09-22 2006-03-24 France Telecom PREEMPTION METHOD FOR MANAGING RADIO RESOURCES IN A MOBILE COMMUNICATION NETWORK
US8116781B2 (en) * 2005-03-28 2012-02-14 Rockstar Bidco Lp Method and system of managing wireless resources
US8457007B2 (en) * 2006-02-24 2013-06-04 Marvell World Trade Ltd. Global switch resource manager
US7711821B2 (en) * 2006-05-31 2010-05-04 International Business Machines Corporation Multiple resource control-advisor for management of distributed or web-based systems
WO2008010673A1 (en) * 2006-07-19 2008-01-24 Lg Electronics Inc. Method for managing data transfer services on a network
US8761792B2 (en) 2008-03-27 2014-06-24 At&T Mobility Ii Llc Management of preemptable communications resources
US8442203B2 (en) * 2008-03-27 2013-05-14 At&T Mobility Ii Llc Identification and determination of non-preemptable calls
EP2355420B1 (en) * 2010-01-26 2014-11-05 Airbus Defence and Space SAS Service admission path control (SAPC)
CN107239347B (en) * 2017-06-15 2020-08-04 杭州迪普科技股份有限公司 Equipment resource allocation method and device in virtual scene
CN110737806B (en) * 2019-09-09 2023-12-01 贝壳技术有限公司 Global ID generation method and device for business entity

Family Cites Families (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4959648A (en) * 1989-02-10 1990-09-25 Motorola, Inc. Dual dynamic priority control in a selctive call system
JP2860661B2 (en) * 1989-03-14 1999-02-24 国際電信電話 株式会社 ATM switch
US5280483A (en) * 1990-08-09 1994-01-18 Fujitsu Limited Traffic control system for asynchronous transfer mode exchange
JP3241716B2 (en) * 1990-08-31 2001-12-25 株式会社東芝 ATM exchange method
US5321692A (en) * 1991-04-09 1994-06-14 Siemens Aktiengesellschaft Method for checking the admissibility of setting up virtual connections
US5850385A (en) * 1991-09-24 1998-12-15 Kabushiki Kaisha Toshiba Cell loss rate sensitive routing and call admission control method
WO1994011971A1 (en) * 1992-11-16 1994-05-26 Fujitsu Limited Call reception control in atm network
US5357507A (en) * 1993-08-24 1994-10-18 Northern Telecom Limited Fast connection admission control for ATM networks
US5446730A (en) * 1993-09-20 1995-08-29 Motorola, Inc. Method for dynamic connection management in integrated communication networks
US5390184A (en) * 1993-09-30 1995-02-14 Northern Telecom Limited Flexible scheduling mechanism for ATM switches
IT1266895B1 (en) * 1994-07-26 1997-01-21 Cselt Centro Studi Lab Telecom PROCEDURE FOR THE OPTIMUM ALLOCATION OF RESOURCES FOR THE TRANSPORT OF VARIABLE BAND INFORMATION FLOWS ON NETWORKS IN ATM TECHNIQUE, AND NODE
JPH0879373A (en) * 1994-09-08 1996-03-22 Fujitsu Ltd Isdn exchange provided with restriction processing function against call congestion
US5490141A (en) * 1994-09-30 1996-02-06 International Business Machines Corporation System and method for providing SVC service through an ATM network for frame relay DTEs with a terminal adapter
DE19507569C2 (en) * 1995-03-03 1997-02-13 Siemens Ag Circuit arrangement for receiving and forwarding message cells through an ATM communication device
US5666364A (en) * 1996-03-01 1997-09-09 Motorola, Inc. Method for prioritizing services provided by different network entities

Also Published As

Publication number Publication date
WO1998051060A1 (en) 1998-11-12
US6141322A (en) 2000-10-31

Similar Documents

Publication Publication Date Title
US6141322A (en) Method and apparatus for precedence and preemption in ATM connection admission control
US5629928A (en) Dynamic fair queuing to support best effort traffic in an ATM network
US6822966B2 (en) Allocating buffers for data transmission in a network communication device
US5982748A (en) Method and apparatus for controlling admission of connection requests
US6144661A (en) Network node apparatus and virtual connection control method for providing various service attributes in multicast communication
US5881050A (en) Method and system for non-disruptively assigning link bandwidth to a user in a high speed digital network
US7046630B2 (en) Packet switching network, packet switching equipment and network management equipment
US6188698B1 (en) Multiple-criteria queueing and transmission scheduling system for multimedia networks
US6934249B1 (en) Method and system for minimizing the connection set up time in high speed packet switching networks
EP1013049B1 (en) Packet network
US6041038A (en) Packet switching device and cell transfer control method
US6760774B1 (en) Boundary apparatus and method for establishing the network connection using a resource reserving function
JPH1070555A (en) Method for data communication in system having exchange organization and port card
JPH07264190A (en) Call reception method in atm exchange
JP4652494B2 (en) Flow control method in ATM switch of distributed configuration
EP0814583A2 (en) Method and system for minimizing the connection set up time in high speed packet switching networks
US6192049B1 (en) Jitterless and lossless switching fabric with network routing
US6359889B1 (en) Cell switching device for controlling a fixed rate connection
JPH10136023A (en) Packet communication method
EP0884923B1 (en) Packet switching network, packet switching equipment, and network management equipment
JP3597049B2 (en) Route-selection type communication device in integrated network
JPH10229405A (en) Atm switch and call reception priority control method
WO1998037730A1 (en) Resource reservation in atm networks
JP3227133B2 (en) ATM switch
Lindgren et al. Fast Connection Establishment in the DTM Gigabit Network.

Legal Events

Date Code Title Description
FZDE Dead