|Publication number||US20080091870 A1|
|Application number||US 11/948,949|
|Publication date||Apr 17, 2008|
|Filing date||Nov 30, 2007|
|Priority date||Jan 5, 2001|
|Also published as||US7346911, US7784061, US20020097720, US20080089343|
|Publication number||11948949, 948949, US 2008/0091870 A1, US 2008/091870 A1, US 20080091870 A1, US 20080091870A1, US 2008091870 A1, US 2008091870A1, US-A1-20080091870, US-A1-2008091870, US2008/0091870A1, US2008/091870A1, US20080091870 A1, US20080091870A1, US2008091870 A1, US2008091870A1|
|Inventors||Brian Goodman, Ronald Hill, Leonard Jesionowski, Michael McIntosh|
|Original Assignee||International Business Machines Corporation|
|Export Citation||BiBTeX, EndNote, RefMan|
|Referenced by (1), Classifications (16), Legal Events (1)|
|External Links: USPTO, USPTO Assignment, Espacenet|
This application is a continuation of U.S. patent application Ser. No. 09/755,405, filed on Jan. 5, 2001, which is incorporated herein by reference in its entirety.
1. Field of the Invention
The present invention relates to a system, method, and manufacture for enabling communication among nodes in a system.
2. Description of the Related Art
Machines or other devices may be comprised of a main controller node that communicates with specific hardware components using a bus interface, such as the Controller Area Network (CAN) serial bus. Each hardware component, e.g., a motor, etc., would include a CAN controller chip to allow communication with the CAN bus and other CAN devices on the bus. A main system controller can transmit commands and data to devices in the system using the CAN message protocol. This arrangement has been satisfactory when there is no need for complex intercomponent communication, such as the case where a main system controller is the primary unit that manages and controls the electro-mechanical devices attached to the CAN bus.
As the complexity of processes that run within the devices and system controller increase and the communication among such processes within the different devices increase, there is a need in the art for improved techniques for implementing the system components and for providing intercommunication among the processes executing within the system components.
Provided is a method, system, and program for allowing communication among nodes in a system. A request is received in a source node from a source object executing in the source node to send a message to a destination object executing in a destination node. Each node includes a processor capable of multitasking multiple program objects and a communication interface to transmit and receive data with the other nodes. A determination is made in the source node as to whether the destination node and source node are a same node. The message is sent in the source node to the destination object within the source node if the destination node is the source node. If the destination node is not the source node, then the source node transmits the message to the destination node through the communication interface. The destination node sends the message to the destination object within the destination node.
In further embodiments, there is a message queue associated with each object in each node. Sending the message to the destination object comprises invoking an operating system command to transmit the message to the message queue associated with the destination object.
In still further embodiments, transmitting the message to the destination node comprises determining an address of the destination node that addresses the destination node when transmitting messages through the transmission medium. At least one message packet is generated including the message, the destination node address, and an address of the destination object, and the at least one message packet is transmitted to the destination node.
Still further, each node may be associated with one component of a system. In such case, a first node comprises a controller node, a second node comprises a component node that controls an electro-mechanical component of the system, the source object comprises a work management object that manages system commands and the message includes a command to instruct a motion object in the component node to control the electro-mechanical component to perform an operation.
Yet further, each object may be assigned a unique identifier in the system, wherein the unique identifier is used within all nodes to identify the destination object to receive the message. Each node may also be assigned a unique node identifier used within all nodes to identify the destination node to receive the message.
In still further embodiments, a function call may receive the request from the source object to send the message to the destination object, determine whether the destination node is the same node, and then send the message to the destination object or cause the transmittal of the message to the destination node. The function call maintains the object and node identifier assignment. In such cases, the node and object identifier used by each function call in each node may be updated to reflect a later modification to the arrangement of nodes or objects in the system.
The described implementations provide an improved system to enable program objects to communicate with other objects in a system comprised of multiple nodes. The system may include nodes that control particular electro-mechanical components of the system. A source object would send a message to a destination object, and the routing of the message to a local queue for the destination object or to another node is seamlessly handled by a function call available at each node. In this way, the source object does not have to be concerned about the location of the destination object as the routing of the message to the destination object is handled by a function call.
Moreover, in certain implementations, the function call uses a unique identifier for each node and object throughout the whole system when determining how to route a message to a destination object. In such embodiments, any modification to the arrangement of objects and nodes can be easily updated by merely updating the unique object and node address assignment used by the function calls in each node.
Referring now to the drawings in which like reference numbers represent corresponding parts throughout:
In the following description, reference is made to the accompanying drawings which form a part hereof and which illustrate several embodiments of the present invention. It is understood that other embodiments may be utilized and structural and operational changes may be made without departing from the scope of the present invention.
In certain cases, the node system is implemented in a storage library system that may include certain of the storage library components shown in
The term “library element” as used herein refers to any location in the library 2 at which a storage cartridge may be located, e.g., the input/output stations, the storage cells 8 a, b, the drives 10 a, b, and gripper 4.
The gripper assembly 4 may also be equipped with a machine vision system, such as a bar code reader, to read a label on the storage cartridge when the gripper assembly 4 is positioned near another library element.
The host interface 30 provides an interface to host systems 36 a, b, c over a communication line, such as a network, serial interface, Small Computer System Interface (SCSI), etc. Additionally, the host interface 30 may communicate with a separate web server or include an embedded web server to enable communication over a network, e.g., an Intranet, Local Area Network (LAN), the Internet, etc. The host systems 36 a, b, c can communicate commands and receive data from the library 20 through the host interface 30. In further embodiments, the host systems 36 a, b, c may communicate with the library 20 through data storage drives (not shown).
In preferred embodiments, each of the storage library components 24, 26, 28, and 30 include a processor node, an XY processor node 38, accessor processor node 40, operator panel processor node 42, and host communication processor node 44, respectively. Each processor node 38, 40, 42, and 44 comprises a processor, memory, firmware to control the processor, and a port to allow communication with a bus interface 46 through which the processor nodes communicate. The bus interface 46 may comprise a controller area network (CAN) bus known in the art, which is a multi-drop network, having a standard access protocol and wiring standards. In alternative embodiments, the bus 46 may comprise any bus or communication interface known in the art. Each of the processor nodes 38, 40, 42, and 44 may either recognize a message identifier associated with each message on the bus interface 46, in accordance with the CAN protocol, or may be specifically addressed with each message, for example as is known in the SCSI bus standard.
In CAN embodiments, each node 38, 40, 42, and 44 would include a CAN controller chip. Each node 38, 40, 42, and 44 would be assigned a unique CAN identifier and groups of nodes may also be assigned a unique identifier to allow messages to be broadcast to multiple nodes. Such CAN identifier may comprise a network address, unique number, etc. The CAN chip in each node 38, 40, 42, and 44 would be configured to receive messages including an identifier or address assigned to address the particular node 38, 40, 42, and 44 and transmit messages including an identifier or address assigned to one of the other nodes 38, 40, 42, and 44. Details of programming the CAN chip are described in the publication entitled “82527 Serial Communications Controller Architectural Overview (Automotive)”, having Intel order no. 272410-003 (Intel Corporation, January 1996), which publication is incorporated herein by reference in its entirety.
In the described embodiments, the components of the library system 20 function as distributed computing elements, each operating under the control of their respective processor node, which performs system specific operations. In preferred embodiments, the accessor processor node 40 functions as the central processor to receive, queue, execute, or distribute host system 36 a, b, c commands. Thus, the accessor processor node 40 provides central processing facilities, including workflow management and queuing. In preferred embodiments, the host communication processor node 44 receives the host system 36 a, b, c commands through the port and interface electronics provided by the host interface 30 and transfers the commands to the accessor processor node 40 over the bus 46. The accessor processor node 40 can then execute such commands to control the gripper assembly 26 servo electronics to move the gripper assembly, or transfer XY motion commands to the XY processor node 38. The XY processor node 38 executes the commands to control the XY system 24 servo-electronics to move the accessor 22 in an XY direction through the tape library 20 to access a storage cartridge in a drive or storage slot. The host system 36 a, b, c command can instruct the accessor 22 to read the identification label on the storage cartridge at a particular location, access a storage cartridge at one library element (e.g., drive, storage slot, gripper, I/O slot, etc.) and move the storage cartridge to another library element.
In further embodiments, the library system 20 may include redundant instances of the above components 22, 28, and 30 to improve the availability of the system and increase processing capabilities. The co-pending and commonly assigned patent applications “High Availability Work Queuing in an Automated Data Storage Library”, having U.S. application Ser. No. 09/573,530, filed May 19, 2000 and “Automated Data Storage Library Distributed Control System”, having U.S. application Ser. No. 09/573,531, filed May 19, 2000, which are both incorporated herein by reference in their entirety, describe further details of a tape library system including distributed processor nodes to implement the components of the tape library system.
Each node further includes objects which perform operations unique to that node. For instance, the accessor processor node 40 includes a work queue object 56 that queues and executes tape library commands from the host systems 36 a, b, c. The host communication processor node 44 includes a host communication object 58 that manages host system 36 a, b, c (
In preferred embodiments, an object executing in one of the nodes 38, 40, 42, and 44 would call a send message function 64 a, b, c, and d to handle the transmission of a message to a local object or an object at a remote node. Thus, the same function call 64 a, b, c, d is used to transmit a message to any object, whether it is on local or remote node. In this way, message communication for an object is seamless. In preferred embodiments, each object in each node is assigned a unique identifier, referred to as an object identifier. The send message function 64 a, b, c, d includes a global mapping of each object to an object identifier, each node to a node identifier, and each object identifier to a node identifier indicating the node in which the object executes. The unique identifier may comprise a unique address, number, code, etc. Thus, for every object identifier there is an associated node number indicating the node in which the object identified by the object number executes. The same global mapping is used by each send message function 64 a, b, c, d in every node 38, 40, 42, 44.
In preferred embodiments, the CAN objects 52 a, b, c, d are capable of transmitting messages to nodes on the interface 46 using a standard network transmission protocol, such as Transmission Control Protocol/Internet Protocol (TCP/IP), Ethernet protocols, proprietary communication protocols, etc. to communicate over a CAN interface 46. The CAN objects 52 a, b, c, d would include the capability to determine the CAN message addresses that correspond to the destination node 72 specified in the message header 70. The CAN message address is used to address the nodes on the CAN bus 46. The source object can specify a group of destination objects. In such case, the destination node field 72 can specify all or a group of nodes that include the destination objects. The CAN object would then select a CAN message address that addresses all the destination nodes specified in field 72 of the message 70. The send message function 64 a, b, c, d fills in the fields of the message header 70.
In further embodiments, the CAN objects 52 a, b, c, d on each node 38, 40, 42, and 44 can transmit messages or pings to determine the availability of the other nodes and determine the immediate status on the connection to the destination object. In this way, the send message function 64 a, b, c, d can be assured that the destination node is available when transmitting messages through the CAN objects 52 a, b, c, d.
Preferred embodiments provide a technique for allowing objects in a distributed computing environment in a system to communicate seamlessly using a single send message function. This single send message function determines whether to route the message to a local object message queue using a standard RTOS queue send function. If the object is not local, then the preferred embodiment send message function sends the message to an object to handle the transmission of the message over the bus interface.
With the preferred embodiments, if the arrangement of objects is altered, then all that has to be updated is the global mapping used by each send message function call. The updated global mapping indicates a new assignment of object identifiers to objects and association of object identifiers to nodes. For instance, if objects from one or more nodes are later consolidated into fewer nodes, then only the global mapping indicating the assignment of objects to nodes would have to be updated in each node to allow the send message function call to use the new arrangement. Further, the removal or addition of nodes and objects can readily be indicated by updating the global mapping to reflect the new arrangement of nodes and objects.
Further, in preferred embodiments, each of the nodes functions independently of the other. This arrangement is different than many CAN systems where there is a master controller and various slaves devices connected to the CAN bus. With the preferred embodiments, each node is an independent processing unit and the preferred embodiments provide a communication interface to allow processes and objects executing independently on the nodes to seamlessly communicate with objects executing in the same or different nodes in the systems.
Following are some alternative implementations for the preferred embodiments.
The preferred embodiments may be implemented as a method, apparatus or article of manufacture using standard programming and/or engineering techniques to produce software, firmware, hardware, or any combination thereof. The term “article of manufacture” as used herein refers to code or logic implemented in hardware logic (e.g., an integrated circuit chip, Field Programmable Gate Array (FPGA), Application Specific Integrated Circuit (ASIC), etc.) or a computer readable medium (e.g., magnetic storage medium (e.g., hard disk drives, floppy disks, tape, etc.), optical storage (CD-ROMs, optical disks, etc.), volatile and non-volatile memory devices (e.g., EEPROMs, ROMs, PROMs, RAMs, DRAMs, SRAMs, firmware, programmable logic, etc.). Code in the computer readable medium is accessed and executed by a processor. The code in which preferred embodiments are implemented may further be accessible through a transmission media or from a file server over a network. In such cases, the article of manufacture in which the code is implemented may comprise a transmission media, such as a network transmission line, wireless transmission media, signals propagating through space, radio waves, infrared signals, etc. Of course, those skilled in the art will recognize that many modifications may be made to this configuration without departing from the scope of the present invention, and that the article of manufacture may comprise any information bearing medium known in the art.
In preferred embodiments, the processor nodes comprised processors that operated under firmware control. In alternative embodiments, the processor nodes may be implemented as hardware logic, such as logic on an integrated circuit chip, e.g., an ASIC, FPGA, etc.
In preferred embodiments, a CAN object is used to generate a message compatible with the CAN protocol to transmit over a CAN bus, using a CAN messaging protocol and CAN chip. In this way, the CAN object provides the interface between the node and the CAN chip. In alternative embodiments, the bus interface may comprise any bus interface known in the art, e.g., Ethernet, LAN etc., and the CAN object may comprise a network object to transmit the message using a network transmission protocol, e.g., TCP/IP, user datagram protocol (UDP), etc., other than CAN. In still further embodiments, the nodes may communicate using any transmission medium known in the art, including a wireless transmission medium. In such alternative embodiments, each node would include hardware to enable communication through the transmission medium according to a communication system and protocol known in the art.
The preferred logic of
In preferred embodiments, the nodes were used to implement the components of a tape library system. In alternative embodiments, the nodes may be part of any system or machine known in the art, where the components of the machine or system may be implemented as distributed processing nodes that communicate over a bus interface.
The foregoing description of the preferred embodiments of the invention has been presented for the purposes of illustration and description. It is not intended to be exhaustive or to limit the invention to the precise form disclosed. Many modifications and variations are possible in light of the above teaching. It is intended that the scope of the invention be limited not by this detailed description, but rather by the claims appended hereto. The above specification, examples and data provide a complete description of the manufacture and use of the composition of the invention. Since many embodiments of the invention can be made without departing from the spirit and scope of the invention, the invention resides in the claims hereinafter appended.
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US7784061||Nov 30, 2007||Aug 24, 2010||International Business Machines Corporation||System and article of manufacture for communication among nodes in a system|
|International Classification||G06F9/46, G06F13/38, H04L29/08, H04L12/40|
|Cooperative Classification||H04L69/329, H04L67/12, G06F9/547, H04L12/407, H04L12/40, H04L2012/40215|
|European Classification||G06F9/54P, H04L29/08N11, H04L12/40, H04L29/08A7, H04L12/407|
|Apr 23, 2008||AS||Assignment|
Owner name: INTERNATIONAL BUSINESS MACHINES CORPORATION, NEW Y
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:HAWLEY, ALLEN;IMMANENI, PAVAN;REEL/FRAME:020844/0051
Effective date: 20031208