Method to Manage Path Failure Threshold Consensus

• US 20100080117A1
• Filed: 09/30/2008
• Published: 04/01/2010
• Est. Priority Date: 09/30/2008
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Hathorn, Roger G., Coronado, Juan A., Holley, Bret W., Valencia, Clarisa
• Granted Patent: US 8,027,263 B2
• Time in Patent Office: Days
• US Class Current: 370/217
• CPC Class Codes: G06F 11/0727 in a storage system, e.g. in a DASD or network based storage system drivers for digital recording or reproducing units G06F3/06; circuits for error detection or correction within digital recording or reproducing units G11B20/18; for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS], H04L67/1097
• CPC Class Codes: G06F 11/076 by exceeding a count or rate limit, e.g. word- or bit count limit

Abstract

A system for providing hosts with a capability to determine which threshold rule of a plurality of threshold rules to use based upon threshold consensus. For example, the system would address a configuration case of several hosts sharing an output port of a fabric via zoning and that port being connected to a single port of a storage controller. If one host is executing lower priority jobs and its threshold is much higher than another host with higher priority jobs and a lower threshold, and the storage controller recognizes that several hosts are sharing the same storage controller port, the consensus will be to ignore the threshold of the first host and to use the threshold of the second host to prevent performance degradation in the system.

Claims

1. A method to minimize performance degradation during communication path failure in a data processing system, the data processing system comprising a plurality of host computers, a storage controller and a plurality of physical paths in communication with the host computer and the storage controller, the method comprising:

establishing a threshold communication path error rate via a failure threshold command for each of the plurality of host computers;

determining whether the plurality of host computers share a common resource corresponding to respective threshold communication path error rates;

performing a consensus operation on the respective threshold communication path error rates to identify a preferred threshold communication path error rate, the consensus operation enabling avoidance of performance degradation due to conflicting threshold communication path error rates;

determining an (i)th actual communication path error rate for an (i)th physical communication path, wherein said (i)th physical communication path is one of said plurality of physical communication paths in communication with said host computer and said storage controller; and

,

discontinuing use of said (i)th physical communication path if said (i)th actual communication path error rate is greater than said preferred threshold communication path error rate.

2. The method of claim 1 wherein:

each of the plurality of host computers comprises at least one channel path identifier (CHPid); and

,

the failure threshold command for a respective host computer enables provision of path failure threshold rules to determine when a CHPid has reached a failed state condition.

3. The method of claim 2 wherein:

the failure threshold command includes parameters that contain path failure threshold rules; and

,

the storage controller uses the path failure threshold rules to determine when a CHPid has reached a failed state.

4. The method of claim 1 wherein:

the data processing system further comprises a switch fabric; and

,

the shared resources comprise the plurality of host computers sharing an output port of the switch fabric

5. The method of claim 4 wherein:

the plurality of host computers share the output port of the switch fabric via zoning.

6. The method of claim 4 wherein:

the output port is coupled to a single port of the storage controller.

7. The method of claim 1 further comprising:

determining whether certain of the plurality of host computers are executing lower priority operations and certain of the plurality of host computers are executing higher priority operations; and

wherein,

the consensus operation ignores a threshold of the host computer executing lower priority operations and chooses to use a threshold of the host computer executing higher priority operations.

14. The apparatus of claim 7 further comprising:

means for determining whether certain of the plurality of host computers are executing lower priority operations and certain of the plurality of host computers are executing higher priority operations; and

wherein,

the consensus operation ignores a threshold of the host computer executing lower priority operations and chooses to use a threshold of the host computer executing higher priority operations.

8. An apparatus to minimize performance degradation during communication path failure in a data processing system, the data processing system comprising a plurality of host computers, a storage controller and a plurality of physical paths in communication with the host computer and the storage controller, the apparatus comprising:

means for establishing a threshold communication path error rate via a failure threshold command for each of the plurality of host computers;

means for determining whether the plurality of host computers share a common resource corresponding to respective threshold communication path error rates;

means for performing a consensus operation on the respective threshold communication path error rates to identify a preferred threshold communication path error rate, the consensus operation enabling avoidance of performance degradation due to conflicting threshold communication path error rates;

means for determining an (i)th actual communication path error rate for an (i)th physical communication path, wherein said (i)th physical communication path is one of said plurality of physical communication paths in communication with said host computer and said storage controller; and

,

means for discontinuing use of said (i)th physical communication path if said (i)th actual communication path error rate is greater than said preferred threshold communication path error rate.

9. The apparatus of claim 8 wherein:

each of the plurality of host computers comprises at least one channel path identifier (CHPid); and

,

the failure threshold command for a respective host computer enables provision of path failure threshold rules to determine when a CHPid has reached a failed state condition.

10. The apparatus of claim 9 wherein:

the failure threshold command includes parameters that contain path failure threshold rules; and

,

the storage controller uses the path failure threshold rules to determine when a CHPid has reached a failed state.

11. The apparatus of claim 8 wherein:

the data processing system further comprises a switch fabric; and

,

the shared resources comprise the plurality of host computers sharing an output port of the switch fabric

12. The apparatus of claim 11 wherein:

the plurality of host computers share the output port of the switch fabric via zoning.

13. The apparatus of claim 11 wherein:

the output port is coupled to a single port of the storage controller.

15. A data processing system comprising

a plurality of host computers,

a storage controller;

a plurality of physical paths in communication with the plurality of host computers and the storage controller; and

,

a system for minimizing performance degradation during communication path failure in a data processing system, the system comprising instructions for;

establishing a threshold communication path error rate via a failure threshold command for each of the plurality of host computers;

determining whether the plurality of host computers share a common resource corresponding to respective threshold communication path error rates;

performing a consensus operation on the respective threshold communication path error rates to identify a preferred threshold communication path error rate, the consensus operation enabling avoidance of performance degradation due to conflicting threshold communication path error rates;

determining an (i)th actual communication path error rate for an (i)th physical communication path, wherein said (i)th physical communication path is one of said plurality of physical communication paths in communication with said host computer and said storage controller; and

,

discontinuing use of said (i)th physical communication path if said (i)th actual communication path error rate is greater than said preferred threshold communication path error rate.

16. The data processing system of claim 15 wherein:

each of the plurality of host computers comprises at least one channel path identifier (CHPid); and

,

the failure threshold command for a respective host computer enables provision of path failure threshold rules to determine when a CHPid has reached a failed state condition.

17. The data processing system of claim 16 wherein:

the failure threshold command includes parameters that contain path failure threshold rules; and

,

the storage controller uses the path failure threshold rules to determine when a CHPid has reached a failed state.

18. The data processing system of claim 15 wherein:

the data processing system further comprises a switch fabric; and

,

the shared resources comprise the plurality of host computers sharing an output port of the switch fabric

19. The data processing system of claim 18 wherein:

the plurality of host computers share the output port of the fabric via zoning.

20. The data processing system of claim 18 wherein:

the output port is coupled to a single port of the storage controller.

21. The data processing system of claim 15 wherein:

determining whether certain of the plurality of host computers are executing lower priority operations and certain of the plurality of host computers are executing higher priority operations; and

wherein,

the consensus operation ignores a threshold of the host computer executing lower priority operations and chooses to use a threshold of the host computer executing higher priority operations.