Efficient volume replication in a storage system

• US 10,656,850 B2
• Filed: 12/21/2018
• Issued: 05/19/2020
• Est. Priority Date: 10/28/2016
• Status: Active Grant

• Current Assignee: Pure Storage, Inc.
• Original Assignee: Pure Storage, Inc.
• Inventors: Golden, Christopher
• Primary Examiner(s): Doan, Khoa D
• Application Number:
  US16/230,560
• Publication Number:
  US 20190155523A1
• Time in Patent Office: 515 Days
• 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 3/065 Replication mechanisms
• CPC Class Codes: G06F 3/0665 at area level, e.g. provisioning of virtual or logical volumes
• CPC Class Codes: G06F 3/0688 Non-volatile semiconductor memory arrays

Abstract

An indication to replicate, for a second storage resource, a volume that is stored at a first storage resource may be received. A snapshot of the volume that is stored at the second storage resource may be identified. Data with operations to data blocks of the volume may be identified. A sequence number range, containing one or more sequence numbers, associated with the snapshot of the volume that is stored at the second resource may be received. Furthermore, a portion of the data associated with the operations to the data blocks of the volume based on the sequence number range that is associated with the snapshot of the volume that is stored at the second resource may be received to replicate the volume at the second storage resource.

Claims

1. A method comprising:

receiving a request to replicate a volume at a source solid-state storage array to a target solid-state storage array, the volume corresponding to a plurality of snapshots associated with data blocks that are stored at the source solid-state storage array;

identifying a most recent snapshot of the plurality of snapshots corresponding to the volume at the source solid-state storage array that is present at the target solid-state storage array;

identifying metadata organized into a plurality of logical groups, each of the logical groups corresponding to a sequence number range and including a portion of the metadata that is associated with a portion of the data blocks of the volume, wherein each sequence number range comprises a plurality of sequence numbers that are assigned to each operation associated with the source solid-state storage array;

determining a last sequence number associated with the most recent snapshot of the plurality of snapshots that is present at the target solid-state storage array;

identifying, by a processing device, a logical group of the plurality of logical groups with a corresponding sequence number range that includes the last sequence number associated with the most recent snapshot; and

reading the metadata to identify a portion of the data blocks to transmit to the target solid-state storage array to replicate the volume based on a next logical group of the plurality of logical groups with a corresponding sequence number range with higher sequence numbers than the identified logical group.

2. The method of claim 1, wherein the sequence numbers are incremented in response to each operation.

3. The method of claim 1, wherein the reading of the metadata to replicate the volume based on the next logical group comprises:

reading additional metadata based on logical groups of the plurality of logical groups with corresponding sequence number ranges with later sequence numbers than the next logical group.

4. The method of claim 1, wherein the metadata specifies the data blocks of the volume that are associated with an operation, and wherein a first group of the plurality of logical groups with later sequence numbers than a second group of the plurality of logical groups specifies operations that are received after operations associated with the second group.

5. The method of claim 2, further comprising:

receiving identifiers associated with sequence numbers for data blocks specified by the next logical group of the plurality of logical groups of the metadata; and

determining, based on the identifiers, whether the data blocks are associated with another sequence number range with prior values for another logical group of the plurality of logical groups than the sequence number range of the next logical group.

6. The method of claim 5, wherein reading the metadata further comprises:

determining to read a first portion of the metadata of the next logical group for data blocks associated with sequence numbers that are included in the sequence number range of the next logical group; and

determining to not read a second portion of the metadata of the next logical group for data blocks associated with sequence numbers that are included in the another sequence number range with prior values than the sequence number range of the next logical group.

7. The method of claim 6, wherein data blocks specified by the next logical group of the plurality of logical groups are associated with a previous snapshot.

8. A system comprising:

a memory; and

a processing device, operatively coupled with the memory, to;

receive an indication to replicate, for a second storage resource, a volume that is stored at a first storage resource;

identify a snapshot of the volume that is stored at the second storage resource;

identify data associated with operations to data blocks of the volume;

receive a sequence number associated with the snapshot of the volume that is stored at the second resource; and

read a portion of the data associated with the operations to the data blocks of the volume based on the sequence number that is associated with the snapshot of the volume that is stored at the second resource to replicate the volume at the second storage resource, wherein the sequence number is associated with a plurality of sequence numbers, each sequence number being assigned to each of the operations to the data blocks of the volume or each operation associated with the first storage resource.

9. The system of claim 8, wherein the volume corresponds to a plurality of snapshots and the snapshot stored at the second storage resource is a most recent snapshot of the volume that is stored at the second storage resource and the sequence number associated with the snapshot stored at the second resource is associated with a last operation to the data blocks of the volume for the snapshot.

10. The system of claim 8, wherein the sequence numbers are incremented or decremented in response to each operation.

11. The system of claim 8, wherein the data is organized into a plurality of logical groups, and wherein a first group of the plurality of logical groups with later sequence numbers than a second group of the plurality of logical groups specifies operations that are received after operations associated with the second group.

12. The system of claim 9, wherein to read the portion of the data, the processing device is further to:

receive identifiers associated with sequence numbers for data blocks associated with the portion of the data; and

determine, based on the identifiers, whether the data blocks associated with the portion of the data are associated with a prior sequence number than sequence numbers of a sequence number range associated with the portion of the data.

13. The system of claim 12, wherein to read the portion of the data, the processing device is further to:

determine to read a first subset of the portion of the data for data blocks associated with sequence numbers that are included in the sequence number range associated with the portion of the data; and

determining to not read a second subset of the portion of the data metadata for data blocks associated with sequence numbers that are included in the another sequence number range with prior values than the sequence number range associated with the portion of the data.

14. A non-transitory computer readable storage medium storing instructions, which when executed, cause a processing device to:

receive an indication to replicate, for a second storage resource, a volume that is stored at a first storage resource;

identify a snapshot of the volume that is stored at the second storage resource;

identify data associated with operations to data blocks of the volume;

receive a sequence number associated with the snapshot of the volume that is stored at the second resource; and

read, by the processing device, a portion of the data associated with the operations to the data blocks of the volume based on the sequence number that is associated with the snapshot of the volume that is stored at the second resource to replicate the volume at the second storage resource, wherein the sequence number is associated with a plurality of sequence numbers, each sequence number being assigned to each of the operations to the data blocks of the volume or each operation associated with the first storage resource.

15. The non-transitory computer readable storage medium of claim 14, wherein the volume corresponds to a plurality of snapshots and the snapshot stored at the second storage resource is a most recent snapshot of the volume that is stored at the second storage resource and the sequence number associated with the snapshot stored at the second resource is associated with a last operation to the data blocks of the volume for the snapshot.

16. The non-transitory computer readable storage medium of claim 14, wherein the sequence numbers are incremented or decremented in response to each operation.

17. The non-transitory computer readable storage medium of claim 14, wherein the data is organized into a plurality of logical groups, and wherein a first group of the plurality of logical groups with later sequence numbers than a second group of the plurality of logical groups specifies operations that are received after operations associated with the second group.

18. The non-transitory computer readable storage medium of claim 14, wherein to read the portion of the data, the processing device is further to:

receive identifiers associated with sequence numbers for data blocks associated with the portion of the data;

determine, based on the identifiers, whether the data blocks associated with the portion of the data are associated with a prior sequence number than sequence numbers of a sequence number range associated with the portion of the data;

determine to read a first subset of the portion of the data for data blocks associated with sequence numbers that are included in the sequence number range associated with the portion of the data; and

determine to not read a second subset of the portion of the data metadata for data blocks associated with sequence numbers that are included in the another sequence number range with prior values than the sequence number range associated with the portion of the data. of the data.