Snapshot creation workflow

• US 9,740,566 B2
• Filed: 09/29/2015
• Issued: 08/22/2017
• Est. Priority Date: 07/31/2015
• Status: Active Grant

• Current Assignee: NetApp, Inc.
• Original Assignee: NetApp, Inc.
• Inventors: Zheng, Ling, Yang, Long, Patel, Kayuri H., Prakash, Suhas, Kimmel, Jeffrey S., Pundir, Anshul, Rokade, Arun
• Primary Examiner(s): YU, JAE UN
• Application Number:
  US14/869,340
• Publication Number:
  US 20170031769A1
• Time in Patent Office: 693 Days
• Field of Search: 711162
• CPC Class Codes: G06F 11/1076 Parity data used in redundant arrays of independent storages, e.g. in RAID systems
• CPC Class Codes: G06F 11/14 Error detection or correction of the data by redundancy in operation G06F11/16 takes precedence
• CPC Class Codes: G06F 11/1446 Point-in-time backing up or restoration of persistent data
• CPC Class Codes: G06F 11/1471 involving logging of persistent data for recovery
• CPC Class Codes: G06F 2201/84 Using snapshots, i.e. a logical point-in-time copy of the data
• 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 3/0659 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]
• CPC Class Codes: G06F 3/0683 Plurality of storage devices

Abstract

A technique efficiently creates a snapshot for a logical unit (LUN) served by a storage input/output (I/O) stack executing on a node of a cluster that organizes data as extents referenced by keys. In addition, the technique efficiently creates one or more snapshots for a group of LUNs organized as a consistency group (CG) and served by storage I/O stacks executing on a plurality of nodes of the cluster. To that end, the technique involves a plurality of indivisible operations (i.e., transactions) of a snapshot creation workflow administered by a Storage Area Network (SAN) administration layer (SAL) of the storage I/O stack in response to a snapshot create request issued by a host. The SAL administers the snapshot creation workflow by initiating a set of transactions that includes, inter alia, (i) installation of barriers for LUNs (volumes) across all nodes in the cluster that participate in snapshot creation, (ii) creation of point-in-time (PIT) markers to record those I/O requests that are included in the snapshot, and (iii) updating of records (entries) in snapshot and volume tables of a cluster database (CDB).

Claims

1. A method comprising:

receiving a first write request directed towards a first logical unit (LUN), the first write request having a data, a logical block address (LBA) and a length representing an address range of the first LUN, the LBA and the length mapped to a first volume associated with the first LUN, the first write request processed at a storage system having a memory and attached to a storage array;

receiving a second write request directed to the first LUN;

in response to receiving a snapshot request from a host, initiating a first barrier on the first volume such that the second write request is blocked;

flushing the first write request by associating a key with the data;

storing the key in a metadata entry included in a metadata structure, the metadata entry associated with the address range; and

creating a first snapshot of the first volume.

2. The method of claim 1 further comprising:

recording a point-in-time (PIT) marker on a non-volatile storage device, wherein the first write request is received before the PIT marker is recorded and the second write request is received after the PIT marker is recorded;

removing the first barrier from the first volume; and

processing the second write request.

3. The method of claim 2 wherein the first and second write requests are associated with first and second sequence numbers respectively, wherein the PIT marker is associated with the first sequence number.

4. The method of claim 1 wherein creating the first snapshot of the first volume further comprises:

sharing the metadata entry between the first volume and a copy of the first volume by increasing a reference count associated with the metadata entry.

5. The method of claim 1 further comprising:

determining whether to coalesce the first and second write requests; and

in response initiating the first barrier, determining not to coalesce the first and second write requests.

6. The method of claim 2 further comprising:

recording the first and second write requests in a replay log;

recording a record in the replay log corresponding to the recording of the PIT marker; and

in response to a crash of the storage system, ignoring replay of the first write request for the first LUN.

7. The method of claim 2 wherein the PIT marker includes a snapshot identifier and a volume identifier.

8. The method of claim 1 further comprising:

creating a group including the first LUN and a second LUN, wherein the second LUN is associated with a second volume; and

in response to the snapshot request from the host directed towards the group, initiating a second barrier on the second volume and creating a second snapshot of the second volume.

9. The method of claim 1 wherein the second write request is not blocked for more than a predetermined period.

10. A system comprising:

a cluster having a plurality of nodes;

a storage array coupled to the cluster;

a first instance of a storage I/O stack executing on a first node of the cluster, the first instance of the storage I/O stack configured to;

receive a first write request directed towards a first logical unit (LUN), the first write request having a data, a logical block address (LBA) and a length representing an address range of the first LUN, the LBA and the length mapped to a volume associated with the first LUN;

a second instance of the storage I/O stack executing on a second node of the cluster, the second instance of the storage I/O stack when configured to;

receive a second write request directed to a second LUN;

associate the first and second LUNs with a group;

in response to receiving a snapshot request for the group from a host, send a barrier message to the first node to establish a first barrier on the volume and initiating a second barrier on the volume at the second node;

determine whether the first and second barriers are established; and

in response to determining that the first and second barriers are established, create snapshots of the first and second LUNs.

11. A system comprising:

a storage system having a memory connected to a processor via a bus;

a storage array coupled to the storage system;

a storage I/O stack executing on the processor of the storage system, the storage I/O stack configured to;

receive a first write request directed towards a first logical unit (LUN), the first write request having a data, a logical block address (LBA) and a length representing an address range of the first LUN, the LBA and the length mapped to a first volume associated with the first LUN;

receive a second write request directed to the first LUN;

in response to receiving a snapshot request from a host, initiate a first barrier on the first volume such that the second write request is blocked;

flush the first write request by associating a key with the data;

store the key in a metadata entry included in a metadata structure, the metadata entry associated with the address range; and

create a first snapshot of the first volume.

12. The system of claim 11 wherein the storage I/O stack is further configured to:

record a point-in-time (PIT) marker on a non-volatile storage device, wherein the first write request is received before the PIT marker is recorded and the second write request is received after the PIT marker is recorded;

remove the first barrier from the first volume; and

process the second write request.

13. The system of claim 12 wherein the first and second write requests are associated with first and second sequence numbers respectively, wherein the PIT marker is associated with the first sequence number.

14. The system of claim 11 wherein the storage I/O stack configured to create the first snapshot of the first volume is further configured to:

share the metadata entry between the first volume and a copy of the first volume by increasing a reference count associated with the metadata entry.

15. The system of claim 11 wherein the storage I/O stack is further configured to:

determine whether to coalesce the first and second write requests; and

in response initiating the first barrier, determine not to coalesce the first and second write requests.

16. The system of claim 12 wherein the storage I/O stack is further configured to:

record the first and second write requests in a replay log;

record a record in the replay log corresponding to the recording of the PIT marker; and

in response to a crash of the storage system, ignore replay of the first write request for the first LUN.

17. The system of claim 12 wherein the PIT marker includes a snapshot identifier and a volume identifier.

18. The system of claim 11 wherein the storage I/O stack is further configured to:

create a group including the first LUN and a second LUN, wherein the second LUN is associated with a second volume; and

in response to the snapshot request from the host directed towards the group, initiate a second barrier on the second volume and create a second snapshot of the second volume.

19. The system of claim 11 wherein the second write request is not blocked for more than 10 seconds.

20. The system of claim 18 wherein the storage I/O stack is further configured to:

record a point-in-time (PIT) marker on a non-volatile storage device, wherein the PIT delimits I/O requests such that an I/O request received after the PIT is blocked, and wherein the PIT is applied to the first and second volumes.