Performing various operations at the granularity of a consistency group within a cross-site storage solution
Systems and methods are described for efficiently performing various operations at the granularity of a consistency group (CG) within a cross-site storage solution. An example of one of the various operations includes an independent and parallel resynchronization approach that independently brings individual volumes of a CG to a steady state of in-synchronization (InSync), thereby contributing to scalability of CGs by supporting CGs having a large number of member volumes without requiring a change to the resynchronization process. Another example includes preserving dependent write-order consistency when a remote mirror copy goes out-of-synchronization (OOS) for any reason by driving all member volumes OOS responsive to any member volume becoming OOS. Yet another example includes independent creation of snapshots by member volumes to support efficient and on-demand creation by an application of a common snapshots of all or a subset of peered member volumes of a CG with which the application is associated.
1. A method performed by one or more processing resources of a distributed storage system, the method comprising:
maintaining state information regarding a data replication status of a mirror copy of a dataset associated with a local consistency group, wherein the distributed storage system includes a plurality of volumes that are members of the local consistency group and wherein the mirror copy is stored on a corresponding plurality of volumes that are members of a remote consistency group of a remote distributed storage system;
responsive to a predetermined event, after a period of time during which the state information indicated each volume of each volume pair of a plurality of volume pairs including a volume of the plurality of volumes of the local consistency group and a corresponding volume of the remote consistency group was in an out-of-synchronization (OOS) state, creating a baseline snapshot of the dataset in a dependent write-order consistent manner;
transferring the baseline snapshot to the remote distributed storage system; and
bringing the mirror copy into an in-synchronization (InSync) state by, for each volume pair of the plurality of volume pairs, orchestrating a resynchronization of the volume pair independent of any other volume pair of the plurality of volume pairs and in parallel with one or more other volume pairs of the plurality of volume pairs.
2. The method of claim 1 , wherein the predetermined event comprises an availability of all volumes of the corresponding plurality of volumes.
3. The method of claim 1 , wherein the mirror copy is in the InSync state when all volume pairs of the plurality of volume pairs are in the InSync state.
4. The method of claim 1 , further comprising:
during performance of the resynchronization, limiting access to the mirror copy to a last known consistent point for the local consistency group and the remote consistency group; and
after all volume pairs of the plurality of volume pairs have transitioned from the OOS state to the InSync state, permitting full access to the mirror copy.
5. The method of claim 1 , wherein the last known consistent point for the local consistency group and the remote consistency group comprises the baseline snapshot.
6. The method of claim 1 , further comprising responsive to a failure by one of more volume pairs of the plurality of volume pairs to transition from the OOS state to the InSync state after performance of the resynchronization, restarting the resynchronization for only the one or more volume pairs.
7. The method of claim 1 , wherein an ability of a particular volume pair of the plurality of volume pairs to achieve the InSync state is unaffected by a failure by another volume of the plurality of volume pairs to achieve the InSync state.
8. A non-transitory computer-readable storage medium embodying a set of instructions, which when executed by one or more processing resources of a distributed storage system, cause the distributed storage system to:
maintain state information regarding a data replication status of a mirror copy of a dataset associated with a local consistency group, wherein the distributed storage system includes a plurality of volumes that are members of the local consistency group and wherein the mirror copy is stored on a corresponding plurality of volumes that are members of a remote consistency group of a remote distributed storage system;
responsive to a predetermined event, after a period of time during which the state information indicated each volume of each volume pair of a plurality of volume pairs including a volume of the plurality of volumes of the local consistency group and a corresponding volume of the remote consistency group was in an out-of-synchronization (OOS) state, create a baseline snapshot of the dataset in a dependent write-order consistent manner;
transfer the baseline snapshot to the remote distributed storage system; and
bring the mirror copy into an in-synchronization (InSync) state by, for each volume pair of the plurality of volume pairs, orchestrating a resynchronization of the volume pair independent of any other volume pair of the plurality of volume pairs and in parallel with one or more other volume pairs of the plurality of volume pairs.
9. The non-transitory computer-readable storage medium of claim 8 , wherein the predetermined event comprises an availability of all volumes of the corresponding plurality of volumes.
10. The non-transitory computer-readable storage medium of claim 8 , wherein the mirror copy is in the InSync state when all volume pairs of the plurality of volume pairs are in the InSync state.
11. The non-transitory computer-readable storage medium of claim 8 , wherein the instructions further cause the distributed storage system to:
during performance of the resynchronization, limit access to the mirror copy to a last
known consistent point for the local consistency group and the remote consistency group; and
after all volume pairs of the plurality of volume pairs have transitioned from the OOS state to the InSync state, permit full access to the mirror copy.
12. The non-transitory computer-readable storage medium of claim 8 , wherein the last known consistent point for the local consistency group and the remote consistency group comprises the baseline snapshot.
13. The non-transitory computer-readable storage medium of claim 8 , wherein the instructions further cause the distributed storage system to responsive to a failure by one of more volume pairs of the plurality of volume pairs to transition from the OOS state to the InSync state after performance of the resynchronization, restart the resynchronization for only the one or more volume pairs.
14. The non-transitory computer-readable storage medium of claim 8 , wherein an ability of a particular volume pair of the plurality of volume pairs to achieve the InSync state is unaffected by a failure by another volume of the plurality of volume pairs to achieve the InSync state.
15. A distributed storage system comprising:
one or more processing resources; and
one or more non-transitory computer-readable media, coupled to the one or more processing resources, having stored therein instructions that when executed by the one or more processing resources cause the distributed storage system to:
maintain state information regarding a data replication status of a mirror copy of a dataset associated with a local consistency group, wherein the distributed storage system includes a plurality of volumes that are members of the local consistency group and wherein the mirror copy is stored on a corresponding plurality of volumes that are members of a remote consistency group of a remote distributed storage system;
responsive to a predetermined event, after a period of time during which the state information indicated each volume of each volume pair of a plurality of volume pairs including a volume of the plurality of volumes of the local consistency group and a corresponding volume of the remote consistency group was in an out-of-synchronization (OOS) state, create a baseline snapshot of the dataset in a dependent write-order consistent manner;
transfer the baseline snapshot to the remote distributed storage system; and
bring the mirror copy into an in-synchronization (InSync) state by, for each volume pair of the plurality of volume pairs, orchestrating a resynchronization of the volume pair independent of any other volume pair of the plurality of volume pairs and in parallel with one or more other volume pairs of the plurality of volume pairs.
16. The distributed storage system of claim 15 , wherein the predetermined event comprises an availability of all volumes of the corresponding plurality of volumes.
17. The distributed storage system of claim 15 , wherein the mirror copy is in the InSync state when all volume pairs of the plurality of volume pairs are in the InSync state.
18. The distributed storage system of claim 15 , wherein the instructions further cause the distributed storage system to:
during performance of the resynchronization, limit access to the mirror copy to a last known consistent point for the local consistency group and the remote consistency group; and
after all volume pairs of the plurality of volume pairs have transitioned from the OOS state to the InSync state, permit full access to the mirror copy.
19. The distributed storage system of claim 15 , wherein the last known consistent point for the local consistency group and the remote consistency group comprises the baseline snapshot.
20. The distributed storage system of claim 15 , wherein the instructions further cause the distributed storage system to responsive to a failure by one of more volume pairs of the plurality of volume pairs to transition from the OOS state to the InSync state after performance of the resynchronization, restart the resynchronization for only the one or more volume pairs.
21. The distributed storage system of claim 15 , wherein an ability of a particular volume pair of the plurality of volume pairs to achieve the InSync state is unaffected by a failure by another volume of the plurality of volume pairs to achieve the InSync state.