Snapshot coordination

1. A method, performed by one or more computing devices, comprising:

handling by a SAN array one or more control plane tasks related to connectivity and one or more snapshots;

creating a logical unit number (LUN) on the SAN array;

coupling communicatively a host cache to the SAN array, wherein non-data network communications between the SAN array and the host cache are used to manage the connectivity and coordinate one or more snapshot requests, wherein the non-data network communications comprises a virtual IP address interface;

creating a cached LUN at the host cache, wherein the cached LUN represents the caching instance of the LUN across a plurality of servers;

receiving at the host cache, a first request to prepare a volume of the host cache for creating a snapshot of the cached LUN wherein the first request indicates that a snapshot of the cached LUN will be taken;

preparing, in response to the first request, the volume of the host cache for creating the snapshot of the cached LUN, wherein preparing the volume of the host cache is dependent on a mode of the host cache, wherein the mode of the host cache comprises at least one of write-through (WT), write-back (WB), or write-back-write-through (WB-WT) switching, and wherein preparing the volume comprises:

flushing a set of data comprising at least one snapshot of the cached LUN when the host cache is in WB or WB-WT switching modes;

switching the host cache to WT mode when the host cache is in WB-WT switching mode; and

sending a message indicating that the volume is prepared for creating the snapshot of the cached LUN;

receiving, at the host cache, a second request to create the snapshot of the cached LUN; in response to the second request, creating, at the host cache, the snapshot of the cached LUN;

switching the host cache to WB mode when the host cache is in WB-WT switching mode;

sending by the host cache a flush complete message to the SAN array to complete movement of the snapshot to the SAN array and to complete creation of the snapshot; and

deleting by the cached LUN, the snapshot.

2. The method of claim 1 , wherein the snapshot comprises a data map of the cached LUN that indicates which data blocks are associated with which files.

3. The method of claim 1 , further comprising receiving a third request to cancel creating the snapshot when preparing the volume for creating the snapshot of the cached LUN and creating the snapshot of the cached LUN exceeds a set amount of time.

4. The method of claim 1 , wherein the preparing includes a timeout value that the cached LUN uses to wait for a freeze request.

5. The method of claim 4 , wherein the flush complete message arrives at the SAN array before a response to the snapshot freeze request is received.

6. The method of claim 1 , wherein creating the snapshot of the cached LUN comprises: holding the input/output (I/O) during the creation of the snapshot; when one or more applications are running, holding the data associated with each of the one or more applications that are running; and generating the snapshot of the cached LUN.

7. The method of claim 1 , further comprising storing at the host cache, the snapshot of the cached LUN when the host cache is in a full time write-back mode.

8. A system comprising:

a SAN array, wherein the SAN array handles one or more control plane tasks related to connectivity and one or more snapshots;

a memory comprising instructions executable by one or more processors; and

the one or more processors coupled to the memory and operable to execute the instructions, the one or more processors being operable when executing the instructions to:

create a logical unit number (LUN) on the SAN array;

communicatively couple a host cache to the SAN array, wherein non-data network communications between the SAN array and the host cache are used to manage the connectivity and coordinate one or more snapshot requests, wherein the non-data network communications comprises a virtual IP address interface;

create a cached LUN at the host cache, wherein the cached LUN represents the caching instance of the LUN across a plurality of servers;

receive at the host cache, a first request to prepare a volume of the host cache for creating a snapshot of the cached wherein the first request indicates that a snapshot of the cached LUN will be taken;

prepare, in response to the first request, the volume of the host cache for creating the snapshot of the cached LUN, wherein preparing the volume of the host cache is dependent on a mode of the host cache, wherein the mode of the host cache comprises at least one of write-through (WT), write-back (WB), or write-back-write-through (WB-WT) switching, and wherein preparing the volume comprises:

flush a set of data comprising at least one snapshot of the cached LUN when the host cache is in WB or WB-WT switching modes;

switch the host cache to WT mode when the host cache is in WB-WT switching mode; and

send a message indicating that the volume is prepared for creating the snapshot of the cached LUN;

receive, at the host cache, a second request to create the snapshot of the cached LUN;

in response to the second request, create, at the host cache, the snapshot of the cached LUN;

switch the host cache to WB mode when the host cache is in WB-WT switching mode;

send, by the host cache, a flush complete message to the SAN array to complete movement of the snapshot to the SAN array and to complete creation of the snapshot; and

delete, by the cached LUN, the snapshot.

9. The system of claim 8 , wherein the snapshot comprises a data map of the cached LUN that indicates which data blocks are associated with which files.

10. The system of claim 8 , further comprising receive a third request to cancel creating the snapshot when preparing the volume for creating the snapshot of the cached LUN and creating the snapshot of the cached LUN exceeds a set amount of time.

11. The system of claim 8 , wherein the prepare includes a timeout value that the cached LUN uses to wait for a freeze request.

12. The system of claim 11 , wherein the flush complete message arrives at the SAN array before a response to the snapshot freeze request is received.

13. The system of claim 8 , wherein creating the snapshot of the cached LUN comprises: hold the input/output (I/O) during the creation of the snapshot; when one or more applications are running, hold the data associated with each of the one or more applications that are running; and generate the snapshot of the cached LUN.

14. The system of claim 8 , further comprising store at the host cache, the snapshot of the cached LUN when the host cache is in a full time write-back mode.

15. One or more computer-readable non-transitory storage media embodying software operable when executed by one or more computer systems to:

handle, by a SAN array, one or more control plane tasks related to connectivity and one or more snapshots;

create a logical unit number (LUN) on the SAN array;

communicatively couple a host cache to the SAN array, wherein non-data network communications between the SAN array and the host cache are used to manage the connectivity and coordinate one or more snapshot requests, wherein the non-data network communications comprises a virtual IP address interface;

create a cached LUN at the host cache, wherein the cached LUN represents the caching instance of the LUN across a plurality of servers;

receive at the host cache, a first request to prepare a volume of the host cache for creating a snapshot of the cached LUN wherein the first request indicates that a snapshot of the cached LUN will be taken;

prepare, in response to the first request, the volume of the host cache for creating the snapshot of the cached LUN, wherein prepare the volume of the host cache is dependent on a mode of the host cache, wherein the mode of the host cache comprises at least one of write-through (WT), write-back (WB), or write-back-write-through (WB-WT) switching, and wherein prepare the volume comprises:

flush a set of data comprising at least one snapshot of the cached LUN when the host cache is in WB or WB-WT switching modes;

switch the host cache to WT mode when the host cache is in WB-WT switching mode; and

send a message indicating that the volume is prepared for creating the snapshot of the cached LUN;

receive, at the host cache, a second request to create the snapshot of the cached LUN;

in response to the second request, create, at the host cache, the snapshot of the cached LUN;

switch the host cache to WB mode when the host cache is in WB-WT switching mode;

send, by the host cache, a flush complete message to the SAN array to complete movement of the snapshot to the SAN array and to complete creation of the snapshot; and

delete, by the cached LUN, the snapshot.

16. The media of claim 15 , wherein the snapshot comprises a data map of the cached LUN that indicates which data blocks are associated with which files.

17. The media of claim 15 , further comprising receive a third request to cancel creating the snapshot when preparing the volume for creating the snapshot of the cached LUN and creating the snapshot of the cached LUN exceeds a set amount of time.

18. The media of claim 15 , wherein the flush complete message arrives at the SAN array before a response to the snapshot freeze request is received.

19. The media of claim 18 , wherein the flush complete message arrives at the SAN array before a response to the snapshot freeze request is received.

20. The media of claim 15 , wherein create the snapshot of the cached LUN comprises:

hold the input/output (I/O) during the creation of the snapshot;

when one or more applications are running, hold the data associated with each of the one or more applications that are running; and

generate the snapshot of the cached LUN.