Dispatching of service requests in redundant storage virtualization subsystems

• US 9,015,391 B2
• Filed: 03/29/2005
• Issued: 04/21/2015
• Est. Priority Date: 03/30/2004
• Status: Active Grant

• Current Assignee: Infortrend Technology Incorporated (Infortrend Corporation)
• Original Assignee: Infortrend Technology Incorporated (Infortrend Corporation)
• Inventors: Schnapp, Michael Gordon, Chan, Chih-Chung
• Primary Examiner(s): Tsai, Henry
• Assistant Examiner(s): ROCHE, JOHN B
• Application Number:
  US10/907,312
• Publication Number:
  US 20050240805A1
• Time in Patent Office: 3,675 Days
• Field of Search: 710/74
• US Class Current: 710/74
• CPC Class Codes: G06F 11/2089 Redundant storage control functionality
• CPC Class Codes: G06F 3/0611 in relation to response time
• CPC Class Codes: G06F 3/0619 in relation to data integrity, e.g. data losses, bit errors
• CPC Class Codes: G06F 3/0635 by changing the path, e.g. traffic rerouting, path reconfiguration
• CPC Class Codes: G06F 3/0665 at area level, e.g. provisioning of virtual or logical volumes
• CPC Class Codes: G06F 3/0683 Plurality of storage devices
• CPC Class Codes: G06F 3/0689 Disk arrays, e.g. RAID, JBOD

Abstract

A method for dispatching of service requests in a redundant storage virtualization subsystem including a plurality of storage virtualization controllers is disclosed. The method comprises the steps of the host entity issuing a first service request to a first one of the SVCs; the first SVC handling the first service request, acquiring a first result of servicing the first service request, and returning the first result to the host entity the result; and, the host entity issuing a second service request to a second one of the SVCs; and, the second SVC handling the second service request, acquiring a second result of servicing the second service request, and returning the second result to the host entity the result. A storage virtualization subsystem implementing the method, a computer system comprising such storage virtualization subsystem, and a storage media having machine-executable codes stored therein for performing the method are also disclosed.

Claims

1. A method for dispatching of service requests in a redundant storage virtualization subsystem (SVS) including a plurality of storage virtualization controllers (SVCs) and a plurality of physical storage devices (PSDs), comprising the steps of:

configuring redundantly a first SVC and a second SVC of the plurality of SVCs such that when one of said first and second SVCs fails, the other of said first and second SVCs will take over functionality of the failed SVC to handle I/O requests;

configuring said first SVC and said second SVC such that not only said first SVC but also said second SVC handle the service requests, and that when the one SVC fails, said the other SVC will take over functionality of the failed SVC to handle said service requests;

mapping, by the plurality of SVCs, combinations of sections of physical storage media of the PSDs into a logical media unit visible to a host entity;

issuing, by said host entity which is external to the plurality of SVCs, a first service request to said first SVC, when said first SVC and said second SVC, which are different SVCs, are both alive;

handling, by said first SVC, said first service request, acquiring a first result of servicing the first service request, and returning said first result to the host entity, when said first SVC and said second SVC are both alive;

issuing, by said host entity, a second service request to said second SVC, when said first SVC and said second SVC are both alive; and

handling, by said second SVC, said second service request, acquiring a second result of servicing the second service request, and returning said second result to the host entity, when said first SVC and said second SVC are both alive;

wherein the service request is selected from one of a group comprising a configuration request, a management request and a monitoring request, and the monitoring request is issued for monitoring not including for IO error monitoring.

2. The method defined in claim 1, wherein said second service request is issued after said first result is returned to the host entity.

3. The method defined in claim 1, wherein said second service request is issued before said first result is returned to the host entity.

4. A method for dispatching of service requests in a redundant storage virtualization subsystem (SVS) including a plurality of storage virtualization controllers (SVCs) and a plurality of physical storage devices (PSDs), comprising the steps of:

configuring redundantly the plurality of SVCs such that when one of the plurality of SVCs fails, another of the plurality of SVCs will take over functionality of the failed SVC to handle I/O requests;

configuring the plurality of SVCs such that the plurality of SVCs handle said service requests, and that when the one SVC fails, said another SVC will take over functionality of the failed SVC to handle said service requests;

mapping, by the plurality of SVCs, combinations of sections of physical storage media of the PSDs into a logical media unit visible to a host entity;

selecting at least one of the SVCs by said host entitywhich is external to the plurality of SVCs, when said plurality of SVCs are alive;

enabling a port on the at least one selected SVC for dispatching of service requests, when said plurality of SVCs are alive;

issuing at least one service request to said at least one selected SVC, when said plurality of SVCs, which are different SVCs, are alive;

handling said at least one service request by said at least one selected SVC, when said plurality of SVCs are alive;

acquiring a result of servicing the at least one service request by said at least one selected SVC, when said plurality of SVCs are alive; and

,

returning to the host entity the result by said at least one selected SVC, when said plurality of SVCs are alive;

wherein the service request is selected from one of a group comprising a configuration request, a management request and a monitoring request, and the monitoring request is issued for monitoring not including for IO error monitoring.

5. The method defined in claim 4, wherein said host entity selects only one of said SVCs to issue said service requests at a time, and said host entity will not issue said service requests to other SVCs until all service requests issued to said one SVC have been serviced to completion.

6. The method defined in claim 5, wherein said selected SVC handles said at least one service request by servicing the service requests.

7. The method defined in claim 4, wherein

said selected SVC handles said at least one service request by forwarding said service requests to an owner of said service request servicing function;

and,

said selected SVC acquires a result of servicing the at least one service request by receiving a service response from said owner.

8. The method defined in claim 4, wherein said step of

said selected SVC handling said at least one service request comprises:

said selected SVC checking if it is an owner of said service request servicing function; and

,

if it is not, then said selected SVC conveying a request for ownership transfer to itself, and said selected SVC servicing the service requests after said selected SVC becomes an owner of said service request servicing function.

9. A method for dispatching of service requests in a redundant storage virtualization subsystem (SVS) including a plurality of storage virtualization controllers (SVCs) and a plurality of physical storage devices (PSDs), comprising the steps of:

configuring redundantly a first SVC and a second SVC of the plurality of SVCs such that when one of said first and second SVCs fails, the other of said first and second SVCs will take over functionality of the failed SVC to handle I/O requests;

configuring said first SVC and said second SVC such that not only said first SVC but also said second SVC may handle said service requests, and that when the one SVC fails, said the other SVC will take over functionality of the failed SVC to handle said service requests;

mapping, by the plurality of SVCs, combinations of sections of physical storage media of the PSDs into a logical media unit visible to a host entity;

selecting, by said host entity which is external to the plurality of SVCs, said first SVC to issue said service requests, when said first SVC and said second SVC, which are different SVCs, are both alive;

receiving, by said first SVC, said service requests, when said first SVC and said second SVC are both alive;

completing, by said first SVC, the servicing of all the received service requests, when said first SVC and said second SVC are both alive; and

,

selecting, by said host entity, said second SVC to issue said service requests, when said first SVC and said second SVC are both alive;

wherein the service request is selected from one of a group comprising a configuration request, a management request and a monitoring request, and the monitoring request is issued for monitoring not including for IO error monitoring.

10. A method for dispatching of service requests in a redundant storage virtualization subsystem (SVS) including a plurality of storage virtualization controllers (SVCs) and a plurality of physical storage devices (PSDs), comprising the steps of:

configuring redundantly a first SVC and a second SVC of the plurality of SVCs such that when one of said first and second SVCs fails, the other of said first and second SVCs will take over functionality of the failed SVC to handle I/O requests;

configuring said first SVC and said second SVC such that not only said first SVC but also said second SVC may handle the service requests, and that when the one SVC fails, said the other SVC will take over functionality of the failed SVC to handle the service requests;

mapping, by the plurality of SVCs, combinations of sections of physical storage media of the PSDs into a logical media unit visible to a host entity;

receiving, by said first SVC, a service request from said host entity which is external to the plurality of SVCs, and forwarding said received service request to said second SVC, when said first SVC and said second SVC, which are different SVCs, are both alive;

dispatching, by said second SVC, the service request for servicing, when said first SVC and said second SVC are both alive;

on completion of request servicing, said second SVC conveying response data and status information to said first SVC, when said first SVC and said second SVC are both alive; and

,

forwarding, by said first SVC, said response data and status information to the host entity, when said first SVC and said second SVC are both alive;

wherein the service request is selected from one of a group comprising a configuration request, a management request and a monitoring request, and the monitoring request is issued for monitoring not including for IO error monitoring.

11. The method defined in claim 10, wherein when said second SVC is in charge of the dispatching of the service requests, any information associated with said service request conveyed between said host entity and said second SVC will be first conveyed to said first SVC and then forwarded by said first SVC to a destination device.

12. A method for dispatching of service requests in a redundant storage virtualization computer system including a plurality of storage virtualization controllers (SVCs) and a plurality of physical storage devices (PSDs), comprising the steps of:

configuring redundantly a plurality of SVCs such that when one of the plurality of SVCs fails, another of the plurality of SVCs will take over functionality of the failed SVC to handle I/O requests;

configuring the plurality of SVCs such that the plurality of SVCs handle said service requests, and that when the one SVC fails, said another SVC will take over functionality of the failed SVC to handle the service requests;

mapping, by the plurality of SVCs, combinations of sections of physical storage media of the PSDs into a logical media unit visible to a host entity;

selecting, by said host entity which is external to the plurality of SVCs, only one of the plurality of SVCs at a time to issue said service requests, when said plurality of SVCs, which are different SVCs, are alive; and

receiving, by said selected SVC, said service requests from said host entity and servicing them, then returning the results of the servicing of the service requests to said host entity, when said plurality of SVCs are alive;

wherein the service request is selected from one of a group comprising a configuration request, a management request and a monitoring request, and the monitoring request is issued for monitoring not including for IO error monitoring.

13. The method defined in claim 12, wherein said host entity checks if all service requests issued to one of said SVCs have been serviced to completion before said selecting one of said SVCs to issue said service requests.

14. The method defined in claim 12, wherein if said issued service requests have not been serviced to completion, said host entity selects one of said SVCs to issue said service requests after said issued service requests have been serviced to completion.

15. The method defined in claim 12, wherein if said issued service requests have not been serviced to completion, said host entity selects said one SVC to issue said service request.

16. A method for dispatching of service requests, said method being performed by a host in a redundant storage virtualization computer system including a plurality of storage virtualization controllers (SVCs) and a plurality of physical storage devices (PSDs), comprising the steps of:

configuring redundantly a plurality of SVCs such that when one of the plurality of SVCs fails, another of the plurality of SVCs will take over functionality of the failed SVC to handle I/O requests;

configuring the plurality of SVCs such that the plurality of SVCs handle said service requests, and that when the one SVC fails, said another SVC will take over functionality of the failed SVC to handle said service requests;

mapping, by the plurality of SVCs, combinations of sections of physical storage media of the PSDs into a logical media unit visible to the host entity;

generating a service request by the host entity which is external to the plurality of redundant SVCs, when the plurality of SVCs, which are different SVCs, are alive;

checking if all issued service requests have been serviced to completion, when the plurality of SVCs are alive;

issuing said service request to one of the redundant SVCs, if all issued service requests have been serviced to completion, when the plurality of SVCs are alive;

otherwise, issuing, by said host entity, said service request to one of said SVCs after said issued service requests have been serviced to completion, when the plurality of SVCs are alive;

wherein the service request is selected from one of a group comprising a configuration request, a management request and a monitoring request, and the monitoring request is issued for monitoring not including for IO error monitoring.

17. The method defined in claim 16, wherein said host entity randomly selects any one of said SVCs as a receiving object.

18. A method for dispatching of service requests, said method being performed by a host in a redundant storage virtualization computer system having a plurality of storage virtualization controllers (SVCs) and a plurality of physical storage devices (PSDs), said SVCs being redundantly configured such that when one of the plurality of SVCs fails, another of the plurality of SVCs will take over functionality of the failed SVC to handle I/O requests, comprising the steps of:

configuring the plurality of SVCs such that the plurality of SVCs handle service requests, and that when the one SVC fails, said another SVC will take over functionality of the failed SVC to handle the service requests;

mapping, by the plurality of SVCs, combinations of sections of physical storage media of the PSDs into a logical media unit visible to the host entity;

generating a new service request by the host which is external to the plurality of SVCs, when the plurality of SVCs, which are different SVCs, are alive;

checking if all previously issued service requests have been serviced to completion, when the plurality of SVCs are alive;

issuing said new service request to one of the SVCs, if all the previously issued service requests have been serviced to completion, when the plurality of SVCs are alive;

otherwise, issuing, by said host entity, said new service request to the SVC which received said service requests issued previously and not serviced to completion, when the plurality of SVCs are alive;

wherein the service request is selected from one of a group comprising a configuration request, a management request and a monitoring request, and the monitoring request is issued for monitoring not including for IO error monitoring.

19. The method defined in claim 18, wherein said step of issuing said service request if all issued service requests have been serviced to completion further comprises the step of said host entity randomly selecting any one of said SVCs as a receiving object of the new service request.

20. A method for dispatching of service requests in a redundant storage virtualization computer system having a plurality of storage virtualization controllers (SVCs) and a plurality of physical storage devices (PSDs), said SVCs being redundantly configured such that when one of the plurality of SVCs fails, another of the plurality of SVCs will take over functionality of the failed SVC to handle I/O requests, comprising the steps of:

configuring the plurality of SVCs such that the plurality of SVCs handle the service requests, and that when the one SVC fails, said another SVC will take over functionality of the failed SVC to handle the service requests;

mapping, by the plurality of SVCs, combinations of sections of physical storage media of the PSDs into a logical media unit visible to a host entity;

selecting, by said host entity which is external to the plurality of SVCs, one of the SVCs to issue the service requests, when the plurality of SVCs, which are different SVCs, are alive;

receiving the service requests from said host entity by said selected SVC and checking if said selected SVC itself is an owner of the service request servicing function, when the plurality of SVCs are alive; and

,

if it is, then said selected SVC servicing the service request, when the plurality of SVCs are alive; and

,

if it is not, said selected SVC forwarding the service requests to an owner of the service request servicing function, when the plurality of SVCs are alive, then said owner receiving and dispatching the service requests for servicing and returning the results of the servicing of the service requests to said host entity, when the plurality of SVCs are alive;

wherein the service request is selected from one of a group comprising a configuration request, a management request and a monitoring request, and the monitoring request is issued for monitoring not including for IO error monitoring.

21. The method defined in claim 20, wherein said owner returns the results of the servicing of the service requests to said host entity through said selected SVC.

22. The method defined in claim 20, further comprising the step of:

any information associated with said service request will be conveyed between said host entity and said owner by first conveying the information to said selected SVC and having said selected SVC forward the information to a destination during said owner servicing the service requests.

23. A method for dispatching of service requests, said method being performed by a first SVC in a redundant storage virtualization computer system having a plurality of storage virtualization controllers (SVCs) and a plurality of physical storage devices (PSDs), said SVCs being redundantly configured such that when one of the plurality of SVCs fails, another of the plurality of SVCs will take over functionality of the failed SVC to handle I/O requests, comprising the steps of:

configuring the plurality of SVCs such that the plurality of SVCs handle the service requests, and that when the one SVC fails, said another SVC will take over functionality of the failed SVC to handle the service requests;

mapping, by the plurality of SVCs, combinations of sections of physical storage media of the PSDs into a logical media unit visible to a host entity;

receiving a service request from said host entity which is external to the plurality of SVCs, when the plurality of SVCs, which are different SVCs, are alive;

forwarding said service request to a second SVC of the plurality of SVCs, when the plurality of SVCs are alive;

receiving response data and status information of servicing said service request from the second SVC, when the plurality of SVCs are alive; and

,

forwarding said data and information to said host entity, when the plurality of SVCs are alive;

wherein the service request is selected from one of a group comprising a configuration request, a management request and a monitoring request, and the monitoring request is issued for monitoring not including for IO error monitoring.

24. The method defined in claim 23, wherein before performing the step of forwarding said service request to a second SVC, said first SVC checks if it is an owner of said service request servicing function, if it is, then said first SVC services said service request without performing the steps of forwarding said service request to the second SVC and receiving response data and status information of servicing said service request from the second SVC.

25. The method defined in claim 24, further comprising the step of, during servicing said service requests, said first SVC conveying information associated with said service request to said host entity and receiving information associated with said service request from said host entity.

26. The method defined in claim 23, further comprising the steps of receiving information associated with said service request from the second SVC and forwarding said information to said host entity, after the step of forwarding said service request to the second SVC.

27. The method defined in claim 23, further comprising the steps of receiving information associated with said service request from said host entity and forwarding said information to the second SVC, after the step of forwarding said service request to the second SVC.

28. A method for dispatching of service requests, said method being performed by a first SVC in a redundant storage virtualization computer system having a plurality of storage virtualization controllers (SVCs) and a plurality of physical storage devices (PSDs), said SVCs being redundantly configured such that when one of the plurality of SVCs fails, another of the plurality of SVCs will take over functionality of the failed SVC to handle I/O requests, comprising the steps of:

configuring the plurality of SVCs such that the plurality of SVCs handle the service requests, and that when the one SVC fails, said another SVC will take over functionality of the failed SVC to handle the service requests;

mapping, by the plurality of SVCs, combinations of sections of physical storage media of the PSDs into a logical media unit visible to a host entity;

receiving a service request from a second SVC of the plurality of SVCs when the plurality of SVCs, which are different SVCs, are alive;

servicing said service request, when the plurality of SVCs are alive; and

,

issuing response data and status information of servicing said service request to the second SVC, when the plurality of SVCs are alive;

wherein the service request is selected from one of a group comprising a configuration request, a management request and a monitoring request, and the monitoring request is issued for monitoring not including for IO error monitoring; and

wherein the service request is issued by said host entity which is external to the plurality of SVCs.

29. The method defined in claim 28, wherein the step of servicing said service request further comprises the steps of:

conveying information associated with said service request to the second SVC; and

, receiving information associated with said service request from the second SVC.

30. The method defined in claim 28, further comprising the steps of:

receiving a service request from the host entity;

servicing said service request from said host entity; and

,

issuing response data and status information of servicing said service request to said host entity.

31. The method defined in claim 30, wherein the step of servicing said service request from said host entity further comprises:

conveying information associated with said service request to said host entity; and

,

receiving information associated with said service request from said host entity.