System and method for write-back cache in sparse volumes

• US 8,601,214 B1
• Filed: 01/06/2011
• Issued: 12/03/2013
• Est. Priority Date: 01/06/2011
• Status: Active Grant

• Current Assignee: NetApp, Inc.
• Original Assignee: NetApp, Inc.
• Inventors: Suggs, Darrell, Chakalian, Mardiros Z., Kumar, Aswini S.
• Primary Examiner(s): Gu, Shawn X
• Application Number:
  US12/986,050
• Time in Patent Office: 1,062 Days
• Field of Search: None
• US Class Current: 711/118
• CPC Class Codes: G06F 12/0804 with main memory updating G06F12/0806 takes precedence
• CPC Class Codes: G06F 12/0873 Mapping of cache memory to specific storage devices or parts thereof
• CPC Class Codes: G06F 16/172 Caching, prefetching or hoarding of files
• CPC Class Codes: G06F 16/1805 Append-only file systems, e.g. using logs or journals to store data
• CPC Class Codes: G06F 2212/264 Remote server
• CPC Class Codes: G06F 3/0659 Command handling arrangements, e.g. command buffers, queues, command scheduling
• CPC Class Codes: G06F 3/067 Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]

Abstract

The techniques introduced here provide for a write-back sparse volume cache in a storage system. The write-back sparse volume cache is implemented by a first storage server that is connected to communicate with a second storage server that hosts the origin storage volume. The first storage server receives a write request from a client to write data to the origin storage volume. In response to receiving the write request, the first storage server writes the data to the sparse volume cache maintained by the first storage server and acknowledges to the client a successful write operation. The data is maintained in the sparse volume cache such that the presence of the data in the sparse volume cache is transparent to the client and subsequent requests for the written data are serviced by the first storage server from the sparse volume cache. The data can later be flushed to the origin storage volume.

Claims

1. A method comprising:

receiving at a first storage server, from a first client, first data to be written to a storage volume on a second storage server, the second storage server communicably coupled with the first storage server;

caching the first data in a sparse volume of the first storage server and indicating to the first client a successful write in response to the first data being cached in the sparse volume, wherein the sparse volume is a further storage volume that appears to the first client to be a complete replication of the storage volume, but does not contain all data from the storage volume;

maintaining client access to the first data in the sparse volume of the first storage server such that the existence of the sparse volume is transparent to the first client; and

sending the first data to the second storage server to be written to the storage volume.

2. The method of claim 1, wherein indicating to the first client a successful write includes sending an acknowledgement without first writing the first data to the storage volume on the second storage server.

3. The method of claim 1, further comprising:

maintaining a data structure on the first storage server; and

adding a first entry to the data structure on the first storage server representing the first data written to the sparse volume, wherein the data structure is a record of data to be written to the storage volume.

4. The method of claim 3, further comprising coalescing entries in the data structure;

and

sending a single write operation for a consecutive range of data blocks represented in the data structure to the second storage server.

5. The method of claim 4, wherein coalescing entries in the data structure includes:

scanning the data structure for a second entry that modifies blocks adjacent to blocks corresponding to the first data; and

combining the second entry with the first entry and removing the second entry from the data structure.

6. The method of claim 3, further comprising collapsing entries in the data structure;

and

sending a single write operation for multiple writes modifying a single set of data represented in the data structure to the second storage server.

7. The method of claim 6, wherein collapsing entries in the data structure includes:

scanning the data structure for a second entry that modifies blocks corresponding to the first data; and

removing the second entry from the data structure.

8. The method of claim 3, further comprising:

stalling new write requests on a defined interval;

flushing all entries in the data structure to the storage volume; and

initiating a backup of the storage volume.

9. The method of claim 1, further comprising providing a second client with access to the first data to be written to the storage volume, on the sparse volume, such that the existence of the sparse volume is transparent to the second client.

10. The method of claim 1, further comprising:

determining if there is a pending operation affecting a file associated with the first data; and

in response to determining that there is a pending operation, suspending caching the first data to the sparse volume of the first storage server until the pending operation is completed.

11. The method of claim 1, wherein sending the first data to the second storage server is in response to receiving at the first storage server, from a second client, second data to be written to the storage volume that overlaps with the first data.

12. The method of claim 1, wherein maintaining the client access to the first data in the sparse volume includes maintaining the client access for the first client.

13. A method comprising:

receiving at a first storage server, from a client, a first set of data to be written to a volume located on a second storage server, wherein the first storage server is connected to the second storage server through a network and the first storage server includes a sparse volume cache of the volume located on the second storage server;

writing the first set of data to the sparse volume cache on the first storage server, wherein the sparse volume cache is a further storage volume that appears to the client to be a complete replication of the volume of the second storage server, but does not contain all data from the volume of the second storage server;

adding a first entry to a data structure on the first storage server, the data structure representing the data written to the sparse volume cache, wherein the data structure is a record of data to be written to the storage volume;

indicating a successful write to the client without first writing the first set of data to the volume located on the second storage server, wherein the presence of the first set of data in the sparse volume cache is transparent to the client; and

sending the data represented in the data structure from the sparse volume cache to the volume located on the second storage server.

14. The method of claim 13, further comprising:

receiving at the first storage server, from the client, a second set of data to be written to the volume on the second storage server, wherein the second set of data is received before the first set of data is sent to the second storage server and the second set of data has an offset range that overlaps with the first set of data;

coalescing entries in the data structure; and

sending a single write operation for a consecutive range of data blocks represented in the data structure to the second storage server.

15. The method of claim 14, wherein coalescing entries in the data structure includes:

scanning the data structure for a second entry corresponding to the second set of data; and

combining the second entry with the first entry and removing the second entry from the data structure.

16. The method of claim 13, further comprising:

receiving at the first storage server, from the client, a second set of data to be written to the volume on the second storage server, wherein the second set of data is received before the first set of data is sent to the second storage server and the second set of data modifies the first set of data;

collapsing entries in the data structure; and

sending a single write operation for multiple writes modifying a single set of data represented in the data structure to the second storage server.

17. The method of claim 16, wherein collapsing entries in the data structure includes:

scanning the data structure for a second entry corresponding to the second set of data; and

removing the first entry from the data structure.

18. The method of claim 13, further comprising providing access for a second client to the sparse volume cache storing the first set of data such that the existence of the sparse volume is transparent to the second client.

19. A storage server comprising:

a processor;

a memory coupled with the processor;

a network adapter configured to allow the storage server to access a second storage server through a network, the second storage server including a storage volume; and

a sparse volume cache of the storage volume, the sparse volume cache configured to receive, from a client, data to be written to the storage volume on the second storage server, wherein in operation, the storage server writes the data to be written to the storage volume, to the sparse volume cache, and sends to the client an indication of a successful write in response to the data being written to the sparse volume cache;

wherein the sparse volume cache is a further storage volume that appears to the client to be a complete replication of the storage volume, but does not contain all data from the storage volume.

20. The storage server of claim 19, further comprising a data structure including entries representing the data written to the sparse volume cache, wherein the data structure is a record of data to be written to the storage volume.

21. The storage server of claim 20, wherein the storage server is configured to send the data represented in the data structure from the sparse volume cache to the storage volume of the second storage server.

22. The storage server of claim 21, wherein the storage server is further configured to coalesce entries in the data structure and send a single write operation for a consecutive range of data blocks to the second storage server.

23. The storage server of claim 21, wherein the storage server is further configured to collapse entries in the data structure and send a single write operation for multiple writes modifying a single data block to the second storage server.

24. The storage server of claim 19, wherein the existence of the sparse volume cache is transparent to the client.

25. The storage server of claim 19, wherein the storage server is configured to provide access to the data written by a first client to the sparse volume cache to a second client such that the presence of the data in the sparse volume cache is transparent to the second client.

26. The storage server of claim 19, further comprising a sent queue configured to maintain write ordering to the storage volume on the second storage server.

27. A storage server comprising;

an interface through which to connect the storage server to communicate with a second storage server, the second storage server including a storage volume;

a sparse volume cache of the storage volume, wherein the sparse volume cache is a further storage volume that appears to a client to be a complete replication of the storage volume, but does not contain all data from the storage volume;

a processor; and

a memory coupled with the processor, the memory storing instructions which when executed by the processor cause the storage server to perform a plurality of operations, including;

receiving, from the client, data to be written to the storage volume;

writing the data from the client to the sparse volume cache;

adding an entry to a data structure stored on the sparse volume cache, the entry representing the data written to the sparse volume cache, wherein the data structure is a record of data to be written to the storage volume;

indicating to the client, in response to the data from the client being written to the sparse volume cache, that the data from the client has been written to the storage volume, wherein the presence of the data from the client in the sparse volume cache is transparent to the client; and

sending the data from the sparse volume cache to the second storage server to be written to the storage volume.

28. The storage server of claim 27, wherein the instructions when executed by the processor further cause the storage server to perform an operation of coalescing entries in the data structure;

and

sending a single write operation for a consecutive range of data blocks represented in the data structure to the second storage server.

29. The storage server of claim 27, wherein the instructions when executed by the processor further cause the storage server to perform an operation of collapsing entries in the data structure;

and

sending a single write operation for multiple writes modifying a single data block to the second storage server.

30. The storage server of claim 27, wherein the instructions when executed by the processor further cause the storage server to perform an operation of providing access to the data written to the sparse volume cache to a second client in response to a request from the second client to access the data on the storage volume, wherein the presence of the data from the client in the sparse volume cache is transparent to the second client.