Methods and systems for implementing shared disk array management functions

• US 7,246,260 B2
• Filed: 03/15/2005
• Issued: 07/17/2007
• Est. Priority Date: 09/24/1998
• Status: Expired due to Term

• Current Assignee: Emc IP Holding Company LLC (Dell Technologies Inc.)
• Original Assignee: Yotta Yotta Incorporated (Dell Technologies Inc.)
• Inventors: Mathews, Michael B., Brown, William P.
• Primary Examiner(s): MASKULINSKI, MICHAEL C
• Application Number:
  US11/082,178
• Publication Number:
  US 20060005076A1
• Time in Patent Office: 854 Days
• Field of Search: 714/5, 714/6, 714/9, 714/11, 714/12
• US Class Current: 714/6.32
• CPC Class Codes: G06F 11/1088 Reconstruction on already foreseen single or plurality of spare disks
• CPC Class Codes: G06F 11/1096 Parity calculation or recalculation after configuration or reconfiguration of the system
• CPC Class Codes: G06F 11/2089 Redundant storage control functionality
• CPC Class Codes: G06F 11/2094 Redundant storage or storage space G06F11/2056 takes precedence
• CPC Class Codes: G06F 11/2097 maintaining the standby controller/processing unit updated initialisation or re-synchronisation thereof G06F11/1658 and subgroups
• CPC Class Codes: G06F 2201/85 Active fault masking without idle spares active fault masking without idle spare hardware where processing functionality is redundant G06F11/2035
• CPC Class Codes: G06F 3/0607 by facilitating the process of upgrading existing storage systems, e.g. for improving compatibility between host and storage device
• CPC Class Codes: G06F 3/061 Improving I/O performance
• CPC Class Codes: G06F 3/0613 in relation to throughput
• CPC Class Codes: G06F 3/0614 Improving the reliability of storage systems
• CPC Class Codes: G06F 3/0619 in relation to data integrity, e.g. data losses, bit errors
• CPC Class Codes: G06F 3/0632 by initialisation or re-initialisation of storage systems
• CPC Class Codes: G06F 3/065 Replication mechanisms
• CPC Class Codes: G06F 3/067 Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]
• CPC Class Codes: G06F 3/0689 Disk arrays, e.g. RAID, JBOD
• CPC Class Codes: G06F 9/526 Mutual exclusion algorithms
• CPC Class Codes: H04L 67/1097 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]

Abstract

Multiple Array Management Functions (AMFs) are connected to multiple redundancy groups over a storage area network (SAN), such as a fiber-channel based SAN. The multiple AMFs share management responsibility of the redundancy groups, each of which typically includes multiple resources spread over multiple disks. The AMFs provide concurrent access to the redundancy groups for associated host systems. When a host requests an AMF to perform an operation on a resource, the AMF synchronizes with the other AMFs sharing control of the redundancy group that includes the resource to be operated on, so as to obtain a lock on the resource. While performing the operation, the AMF send replication data and state information associated with the resource such that if the AMF fails, any of the other AMFs are able to complete the operation and maintain data reliability and coherency.

Claims

1. A data storage system comprising:

a redundancy group including a plurality of resources;

two or more array management functions (AMFs) sharing access to the redundancy group, wherein the AMFs provide concurrent access to the redundancy group for associated host systems; and

an interconnect medium for connecting the AMFs with the redundancy group;

wherein when a first one of the AMFs desires to perform an operation on a first resource in the redundancy group, the first AMF arbitrates with the other AMFs sharing access to the redundancy group for a lock on the first resource, whereupon the first AMF performs the operation on the first resource and concurrently sends replication data and state information associated with the first resource, the replication data and state information including the state of completion of the operation, to the other AMFs such that if the first AMF fails while performing the operation, one of the other AMFs is able to complete the operation and wherein if the first AMF fails while performing the operation, one of the other AMFs completes the operation by using sent replication data and state information.

2. The data storage system of claim 1, wherein the first AMF does not release the lock on the first resource until a second one of the AMFs arbitrates for a lock on the first resource.

3. The data storage system of claim 1, wherein if the first AMF fails, the remaining AMFs arbitrate for a lock on the first resource, whereupon a second one of the AMFs obtains the lock and completes the operation.

4. The data storage system of claim 1, wherein the operation performed by the first AMF on the first resource includes a plurality of steps, wherein the first AMF performs each step of the operation on the resource, and for each step concurrently sends replication data and state information associated with the first resource to the remaining AMFs, such that if the first AMF fails while performing any of the steps of the operation, one of the remaining AMFs is able to complete the operation.

5. The data storage system of claim 1, wherein one or more host systems communicate with the first AMF over one of the interconnect medium, an internet connection and a PCI bus.

6. The system of claim 1, wherein the interconnect medium includes at least one of a SCSI interface, a fiber-channel interface, a storage area network and a fiber-channel storage area network.

7. The system of claim 1, wherein each of the AMFs executes in one of a controller, a host bus adapter and a host computer.

8. A method of dynamically sharing management of a redundancy group in a data storage system between two or more array management functions (AMFs), wherein the AMFs are able to concurrently access the redundancy group, the redundancy group including a plurality of resources, the method comprising the steps of:

receiving a request from a host by a first one of the AMFs to perform a first operation on a first one of the resources;

synchronizing with the other AMFs so as to acquire access to the first resource; and

performing the first operation on the first resource;

sending replication data and state information associated with the first resource to the other AMFs, the replication data and state information including the state of completion of the operation,

concurrently with performing the first operation, such that if the first AMF fails, one of the other AMFs is able to complete the first operation; and

if the first AMF fails while performing the first operation, completing the first operation with a second one of the AMFs.

9. The method of claim 8, wherein the first operation is a read operation, the method further comprising the steps of:

receiving a request from a host by a second one of the AMFs to perform a second read operation on the first resource; and

performing the second read operation on the first resource by the second AMF concurrently with the first AMF performing the first read operation on the first resource.

10. The method of claim 8, wherein the step of synchronizing includes the step of arbitrating with the one or more other AMFs to acquire a lock on the first resource so that no other AMF is able to access the first resource while the first AMF maintains the lock.

11. The method of claim 10, further comprising the step of releasing the lock on the first resource only when another AMF arbitrates for the lock.

12. A method according to claim 8, wherein the step of completing the first operation comprises the step of the second AMF synchronizing with the other AMFs so as to acquire access to the first resource.

13. A method according to claim 8, wherein the first operation includes a plurality of sub-operations, wherein the step of performing the first operation includes performing each sub-operation on the first resource, the method further comprising sending replication data and state information associated with the first resource to the other AMFs concurrently with the performance of each sub-operation, such that if the first AMF fails, one of the other AMFs is able to complete the first operation.

14. A method according to claim 8, further comprising the steps of:

receiving a request from a second host by a second one of the AMFs to perform a second operation on a second one of the resources;

synchronizing with the first AMF and the other AMFs so as to acquire access to the second resource; and

performing the second operation on the second resource.

15. A method according to claim 14, wherein at least a portion of the second operation is performed concurrently with the performance of the first operation.

16. A data storage network system comprising:

one or more redundancy groups, each redundancy group including multiple resources spread over multiple disks;

two or more array management functions (AMFs) sharing redundancy group management of the one or more redundancy groups, wherein the AMFs are able to concurrently access the one or more redundancy groups; and

an interconnect medium for interconnecting the AMFs with the one or more redundancy groups, wherein

when a first one of the AMFs receives a write command from a host to write at least two data sets to two or more of the resources, the first AMF acquires a lock on the first resource to which the first data set is to be written, writes the first data set to the first resource and concurrently performs a replication operation wherein replication data and state information associated with the first resource is sent to the other AMFs, the replication data and state information including the state of completion of the operation, such that if the first AMF fails while performing the write operation, one of the other AMFs is able to complete the write operation; and

wherein, if the first AMF fails while performing the operation, one of the other AMFs completes the operation by using sent replication data and state information.

17. A system according to claim 16, wherein a first one of the AMFs receives a write command from a host to write at least two data sets to two or more of the resources, wherein the first AMF acquires a lock on the first resource to which the first data set is to be written, writes the first data set to the first resource and concurrently performs a replication operation wherein replication data and state information associated with the first resource is sent to the other AMFs, such that if the first AMF fails while performing the write operation, one of the other AMFs is able to complete the write operation.

18. A system according to claim 17, wherein concurrently with sending replication data and state information, the first AMF determines if the second data set is to be written to a second resource, and if so acquires a lock on the second resource, writes the second data set to the second resource and concurrently performs a second replication operation wherein replication data and state information associated with the second resource is sent to the other AMFs, whereupon the first AMF waits until all replication operations have completed before sending status information to the host.

19. A system according to claim 16, wherein each of the AMFs includes a means for pipelining replication of incoming host data.

20. A system according to claim 16, wherein the interconnect medium includes at least one of a SCSI interface, a fiber-channel interface, a storage area network and a fiber-channel storage area network.