Data communications in a parallel active messaging interface of a parallel computer

• US 8,949,453 B2
• Filed: 11/30/2010
• Issued: 02/03/2015
• Est. Priority Date: 11/30/2010
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Archer, Charles J., Blocksome, Michael A., Ratterman, Joseph D., Smith, Brian E.
• Primary Examiner(s): Mesfin, Yemane
• Assistant Examiner(s): Rose, Derrick V
• Application Number:
  US12/956,903
• Publication Number:
  US 20120137294A1
• Time in Patent Office: 1,526 Days
• Field of Search: 370/331, 712/11, 719/313
• US Class Current: 709/232
• CPC Class Codes: G06F 9/546 Message passing systems or structures, e.g. queues

Abstract

Data communications in a parallel active messaging interface (‘PAMI’) of a parallel computer, the PAMI composed of data communications endpoints, each endpoint including a specification of data communications parameters for a thread of execution on a compute node, including specifications of a client, a context, and a task, endpoints coupled for data communications through the PAMI and through data communications resources, including receiving in an origin endpoint of the PAMI a SEND instruction, the SEND instruction specifying a transmission of transfer data from the origin endpoint to a first target endpoint; transmitting from the origin endpoint to the first target endpoint a Request-To-Send (‘RTS’) message advising the first target endpoint of the location and size of the transfer data; assigning by the first target endpoint to each of a plurality of target endpoints separate portions of the transfer data; and receiving by the plurality of target endpoints the transfer data.

Claims

1. A parallel computer that processes data communications in a parallel active messaging interface (‘

PAMI’

) of a parallel computer, the parallel computer comprising a plurality of compute nodes that execute a parallel application, the PAMI comprising data communications endpoints, each endpoint comprising a specification of data communications parameters for a thread of execution on a compute node, including specifications of a client, a context, and a task, the compute nodes and the endpoints coupled for data communications through the PAMI and through data communications resources, the compute nodes comprising computer processors operatively coupled to computer memory having disposed within it computer program instructions that, when executed by the computer processors, cause the parallel computer to function by;

receiving in an origin endpoint of the PAMI a SEND instruction, the SEND instruction specifying a transmission of transfer data from the origin endpoint to a first target endpoint of a plurality of target endpoints;

transmitting from the origin endpoint to the first target endpoint a Request-To-Send (‘

RTS’

) message advising the first target endpoint of the location and size of the transfer data;

assigning by the first target endpoint to each of the plurality of target endpoints separate portions of the transfer data by;

dividing the size of the transfer data by the number of the plurality of target endpoints, and

posting receive instructions in work queues of contexts of each of the plurality of target endpoints, each receive instruction specifying a separate portion of the transfer data by a pointer to a separate sub-portion of a source buffer of the first target endpoint; and

receiving by the plurality of target endpoints the transfer data.

2. The parallel computer of claim 1 wherein receiving the transfer data further comprises:

receiving by each of the plurality of target endpoints the separate portion of the transfer data assigned to that target endpoint.

3. The parallel computer of claim 1 wherein:

the computer memory further has disposed within it computer program instructions that, when executed by the computer processors, cause the parallel computer to function by dividing by the first target endpoint the transfer data into a number of portions that is larger than the number of target endpoints in the plurality of target endpoints;

assigning portions of the transfer data further comprises iteratively assigning by the first target endpoint to each of the plurality of target endpoints additional separate portions of the transfer data until all the transfer data is assigned among the plurality of target endpoints; and

receiving the transfer data further comprises iteratively receiving by each of the plurality of target endpoints all of the additional separate portions of the transfer data assigned to that target endpoint.

4. The parallel computer of claim 1 wherein:

assigning separate portions of the transfer data comprises posting by an advance function of the first target endpoint receive instructions into work queues of contexts of the plurality of target endpoints, each receive instruction specifying a separate portion of the transfer data; and

receiving the transfer data comprises executing, by advance functions of the contexts in the plurality of target endpoints, the receive instructions.

5. The parallel computer of claim 1 wherein

the receiving of the transfer data is implemented by advance functions of contexts in the plurality of endpoints;

and

each advance function in the plurality of target endpoints executes in a separate thread of execution.

6. The parallel computer of claim 1 wherein:

each client comprises a collection of data communications resources dedicated to the exclusive use of an application-level data processing entity;

each context comprises a subset of the collection of data processing resources of a client, context functions, and a work queue of data transfer instructions to be performed by use of the subset through the context functions operated by an assigned thread of execution; and

each task represents a process of execution of the parallel application.

7. A computer program product for processing data communications in a parallel active messaging interface (‘

PAMI’

) of a parallel computer, the parallel computer comprising a plurality of compute nodes that execute a parallel application, the PAMI comprising data communications endpoints, each endpoint comprising a specification of data communications parameters for a thread of execution on a compute node, including specifications of a client, a context, and a task, the compute nodes and the endpoints coupled for data communications through the PAMI and through data communications resources, the computer program product disposed upon a non-transitory computer readable storage medium, the computer program product comprising computer program instructions that, when installed and executed, cause the parallel computer to function by;

receiving in an origin endpoint of the PAMI a SEND instruction, the SEND instruction specifying a transmission of transfer data from the origin endpoint to a first target endpoint of a plurality of target endpoints;

transmitting from the origin endpoint to the first target endpoint a Request-To-Send (‘

RTS’

) message advising the first target endpoint of the location and size of the transfer data;

assigning by the first target endpoint to each of the plurality of target endpoints separate portions of the transfer data by;

dividing the size of the transfer data by the number of the plurality of target endpoints, and posting receive instructions in work queues of contexts of each of the plurality of target endpoints, each receive instruction specifying a separate portion of the transfer data by a pointer to a separate sub-portion of a source buffer of the first target endpoint; and

receiving by the plurality of target endpoints the transfer data.

8. The computer program product of claim 7 wherein receiving the transfer data further comprises:

receiving by each of the plurality of target endpoints the separate portion of the transfer data assigned to that target endpoint.

9. The computer program product of claim 7 wherein:

the computer program product further comprises computer program instructions that, when installed and executed, cause the parallel computer to function by dividing by the first target endpoint the transfer data into a number of portions that is larger than the number of target endpoints in the plurality of target endpoints;

assigning portions of the transfer data further comprises iteratively assigning by the first target endpoint to each of the plurality of target endpoints additional separate portions of the transfer data until all the transfer data is assigned among the plurality of target endpoints; and

receiving the transfer data further comprises iteratively receiving by each of the plurality of target endpoints all of the additional separate portions of the transfer data assigned to that target endpoint.

10. The computer program product of claim 7 wherein

assigning separate portions of the transfer data comprises posting by an advance function of the first target endpoint receive instructions into work queues of contexts of the plurality of target endpoints, each receive instruction specifying a separate portion of the transfer data;

and

receiving the transfer data comprises executing, by advance functions of the contexts in the plurality of target endpoints, the receive instructions.

11. The computer program product of claim 7 wherein

the receiving of the transfer data is implemented by advance functions of contexts in the plurality of endpoints;

and

each advance function in the plurality of target endpoints executes in a separate thread of execution.

12. The computer program product of claim 7 wherein:

each client comprises a collection of data communications resources dedicated to the exclusive use of an application-level data processing entity;

each context comprises a subset of the collection of data processing resources of a client, context functions, and a work queue of data transfer instructions to be performed by use of the subset through the context functions operated by an assigned thread of execution; and

each task represents a process of execution of the parallel application.