Routing Data Communications Packets In A Parallel Computer

• US 20120079133A1
• Filed: 09/28/2010
• Published: 03/29/2012
• Est. Priority Date: 09/28/2010
• Status: Active Grant

• 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., INGLETT, TODD A.
• Granted Patent: US 9,569,398 B2
• Time in Patent Office: Days
• US Class Current: 709/238
• CPC Class Codes: G06F 15/17312 Routing techniques specific to parallel machines, e.g. wormhole, store and forward, shortest path problem congestion routing on a LAN H04L45/00
• CPC Class Codes: G06F 9/45558 Hypervisor-specific management and integration aspects
• CPC Class Codes: H04L 45/52 Multiprotocol routers
• CPC Class Codes: H04L 67/63 Routing a service request depending on the request content or context

Abstract

Routing data communications packets in a parallel computer that includes compute nodes organized for collective operations, each compute node including an operating system kernel and a system-level messaging module that is a module of automated computing machinery that exposes a messaging interface to applications, each compute node including a routing table that specifies, for each of a multiplicity of route identifiers, a data communications path through the compute node, including: receiving in a compute node a data communications packet that includes a route identifier value; retrieving from the routing table a specification of a data communications path through the compute node; and routing, by the compute node, the data communications packet according to the data communications path identified by the compute node'"'"'s routing table entry for the data communications packet'"'"'s route identifier value.

Claims

1. A method of routing data communications packets in a parallel computer, the parallel computer comprising a plurality of compute nodes organized for collective operations through at least one data communications network,

each compute node comprising an operating system kernel and a system-level messaging module, the system-level messaging module comprising a module of automated computing machinery that exposes a messaging interface to applications,

each compute node including a routing table that specifies, for each of a multiplicity of route identifiers, a data communications path through the compute node,

the method comprising:

receiving in a compute node a data communications packet, the data communications packet including a route identifier value that identifies a specification of a data communications path through each compute node that the data communications packet traverses;

retrieving from the routing table in dependence upon the route identifier value the specification of a data communications path through the compute node; and

routing, by the compute node, the data communications packet through the compute node according to the data communications path identified by the compute node'"'"'s routing table entry for the data communications packet'"'"'s route identifier value.

2. The method of claim 1 further comprising:

determining whether the compute node'"'"'s routing table contains an entry for the data communications packet'"'"'s route identifier value; and

creating by the operating system kernel an entry for the data communications packet'"'"'s route identifier value in the compute node'"'"'s routing table if the compute node'"'"'s routing table does not have an entry for the data communications packet'"'"'s route identifier value.

3. The method of claim 1 further comprising initializing by the operating system kernel the compute node'"'"'s routing table with a predefined set of startup routing table entries.

4. The method of claim 1 wherein the compute node'"'"'s routing table includes a maximum number of entries, and the total number of data communications paths through the compute node is greater than the maximum number of entries in the compute node'"'"'s routing table.

5. The method of claim 1 wherein:

the compute node'"'"'s routing table contains a maximum number of entries;

the total number of data communications paths through the compute node is greater than the maximum number of entries in the compute node'"'"'s routing table;

the operating system kernel has a larger routing table containing more entries than the maximum number of entries in the compute node'"'"'s routing table; and

creating by the operating system kernel an entry for the data communications packet'"'"'s route identifier value in the compute node'"'"'s routing table further comprises providing an entry from the larger routing table in the operating system kernel.

6. The method of claim 1 wherein:

the compute node'"'"'s routing table contains a maximum number of entries;

the compute node'"'"'s routing table is full;

creating by the operating system kernel an entry for the data communications packet'"'"'s route identifier value in the compute node'"'"'s routing table further comprises evicting an entry from the compute node'"'"'s routing table according to a predetermined eviction policy and replacing the evicted entry in the compute node'"'"'s routing table with the entry for the data communications packet'"'"'s route identifier value.

7. Apparatus for routing data communications packets in a parallel computer, the parallel computer comprising a plurality of compute nodes organized for collective operations through at least one data communications network,

each compute node comprising an operating system kernel and a system-level messaging module, the system-level messaging module comprising a module of automated computing machinery that exposes a messaging interface to applications,

each compute node including a routing table that specifies, for each of a multiplicity of route identifiers, a data communications path through the compute node,

the apparatus comprising a computer processor, a computer memory operatively coupled to the computer processor, the computer memory having disposed within it computer program instructions that, when executed by the computer processor, cause the apparatus to carry out the steps of:

receiving in a compute node a data communications packet, the data communications packet including a route identifier value that identifies a specification of a data communications path through each compute node that the data communications packet traverses;

retrieving from the routing table in dependence upon the route identifier value the specification of a data communications path through the compute node; and

routing, by the compute node, the data communications packet through the compute node according to the data communications path identified by the compute node'"'"'s routing table entry for the data communications packet'"'"'s route identifier value.

8. The apparatus of claim 7 wherein the computer memory has disposed within it computer program instructions that, when executed by the computer processor, cause the apparatus to carry out the steps of:

determining whether the compute node'"'"'s routing table contains an entry for the data communications packet'"'"'s route identifier value; and

creating by the operating system kernel an entry for the data communications packet'"'"'s route identifier value in the compute node'"'"'s routing table if the compute node'"'"'s routing table does not have an entry for the data communications packet'"'"'s route identifier value.

9. The apparatus of claim 7 wherein the computer memory has disposed within it computer program instructions that, when executed by the computer processor, cause the apparatus to carry out the step of initializing by the operating system kernel the compute node'"'"'s routing table with a predefined set of startup routing table entries.

10. The apparatus of claim 7 wherein the compute node'"'"'s routing table includes a maximum number of entries, and the total number of data communications paths through the compute node is greater than the maximum number of entries in the compute node'"'"'s routing table.

11. The apparatus of claim 7 wherein:

the compute node'"'"'s routing table contains a maximum number of entries;

the total number of data communications paths through the compute node is greater than the maximum number of entries in the compute node'"'"'s routing table;

the operating system kernel has a larger routing table containing more entries than the maximum number of entries in the compute node'"'"'s routing table; and

creating by the operating system kernel an entry for the data communications packet'"'"'s route identifier value in the compute node'"'"'s routing table further comprises providing an entry from the larger routing table in the operating system kernel.

12. The apparatus of claim 7 wherein:

the compute node'"'"'s routing table contains a maximum number of entries;

the compute node'"'"'s routing table is full;

creating by the operating system kernel an entry for the data communications packet'"'"'s route identifier value in the compute node'"'"'s routing table further comprises evicting an entry from the compute node'"'"'s routing table according to a predetermined eviction policy and replacing the evicted entry in the compute node'"'"'s routing table with the entry for the data communications packet'"'"'s route identifier value.

13. A computer program product for routing data communications packets in a parallel computer, the parallel computer comprising a plurality of compute nodes organized for collective operations through at least one data communications network,

each compute node comprising an operating system kernel and a system-level messaging module, the system-level messaging module comprising a module of automated computing machinery that exposes a messaging interface to applications,

each compute node including a routing table that specifies, for each of a multiplicity of route identifiers, a data communications path through the compute node,

the computer program product disposed upon a computer readable storage medium, the computer program product comprising computer program instructions that, when executed, cause the parallel computer to carry out the steps of:

receiving in a compute node a data communications packet, the data communications packet including a route identifier value that identifies a specification of a data communications path through each compute node that the data communications packet traverses;

retrieving from the routing table in dependence upon the route identifier value the specification of a data communications path through the compute node; and

routing, by the compute node, the data communications packet through the compute node according to the data communications path identified by the compute node'"'"'s routing table entry for the data communications packet'"'"'s route identifier value.

14. The computer program product of claim 13 further comprising computer program instructions that, when executed, cause the parallel computer to carry out the steps of:

determining whether the compute node'"'"'s routing table contains an entry for the data communications packet'"'"'s route identifier value; and

creating by the operating system kernel an entry for the data communications packet'"'"'s route identifier value in the compute node'"'"'s routing table if the compute node'"'"'s routing table does not have an entry for the data communications packet'"'"'s route identifier value.

15. The computer program product of claim 13 further comprising computer program instructions that, when executed, cause the parallel computer to carry out the step of initializing by the operating system kernel the compute node'"'"'s routing table with a predefined set of startup routing table entries.

16. The computer program product of claim 13 wherein the compute node'"'"'s routing table includes a maximum number of entries, and the total number of data communications paths through the compute node is greater than the maximum number of entries in the compute node'"'"'s routing table.

17. The computer program product of claim 13 wherein:

the compute node'"'"'s routing table contains a maximum number of entries;

the total number of data communications paths through the compute node is greater than the maximum number of entries in the compute node'"'"'s routing table;

the operating system kernel has a larger routing table containing more entries than the maximum number of entries in the compute node'"'"'s routing table; and

creating by the operating system kernel an entry for the data communications packet'"'"'s route identifier value in the compute node'"'"'s routing table further comprises providing an entry from the larger routing table in the operating system kernel.

18. The computer program product of claim 13 wherein:

the compute node'"'"'s routing table contains a maximum number of entries;

the compute node'"'"'s routing table is full;

creating by the operating system kernel an entry for the data communications packet'"'"'s route identifier value in the compute node'"'"'s routing table further comprises evicting an entry from the compute node'"'"'s routing table according to a predetermined eviction policy and replacing the evicted entry in the compute node'"'"'s routing table with the entry for the data communications packet'"'"'s route identifier value.