Concurrent transactional checkpoints in a clustered file system

• US 8,516,020 B2
• Filed: 08/06/2012
• Issued: 08/20/2013
• Est. Priority Date: 08/25/2008
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Aronovich, Lior, Toaff, Yair, Paz, Gil, Asher, Ron
• Primary Examiner(s): Casanova, Jorge A
• Application Number:
  US13/567,228
• Publication Number:
  US 20120303683A1
• Time in Patent Office: 379 Days
• Field of Search: 707/786, 707/826
• US Class Current: 707/826
• CPC Class Codes: G06F 12/0802 Addressing of a memory level in which the access to the desired data or data block requires associative addressing means, e.g. caches
• CPC Class Codes: G06F 12/0815 Cache consistency protocols
• CPC Class Codes: G06F 12/0828 with concurrent directory accessing, i.e. handling multiple concurrent coherency transactions
• CPC Class Codes: G06F 12/084 with a shared cache
• CPC Class Codes: G06F 16/11 File system administration, e.g. details of archiving or snapshots file system backup G06F11/14
• CPC Class Codes: G06F 16/13 File access structures, e.g. distributed indices arrangements of input from, or output to, record carriers G06F3/06
• CPC Class Codes: G06F 16/16 File or folder operations, e.g. details of user interfaces specifically adapted to file systems
• CPC Class Codes: G06F 16/172 Caching, prefetching or hoarding of files
• CPC Class Codes: G06F 16/182 Distributed file systems
• CPC Class Codes: G06F 16/183 Provision of network file services by network file servers, e.g. by using NFS, CIFS network file access protocols H04L67/1097
• CPC Class Codes: G06F 16/1865 Transactional file systems
• CPC Class Codes: G06F 16/24552 Database cache management
• CPC Class Codes: G06F 21/6218 to a system of files or objects, e.g. local or distributed file system or database
• CPC Class Codes: G06F 2212/604 Details relating to cache allocation

Abstract

Systems, Methods, and Computer Program Products are provided for performing concurrent checkpoints from file system agents residing on different nodes within in a clustered file system (CFS). Responsibility to checkpoint a modified and a committed data segment to a final storage location is assigned to one of the file system agents. One of the file system agents, which is assigned, is the file system agent whose associated distributed shared memory (DSM) agent is an owner of the data segment.

Claims

1. A method for performing concurrent transactional checkpoints from a plurality of file system agents residing on different nodes within in a clustered file system (CFS) using a processor device, the method comprising:

assigning to one of the plurality of file system agents responsibility to checkpoint a modified data segment and a committed data segment to a final storage location, wherein the one of the plurality of file system agents assigned is the one of the plurality of file system agents whose associated distributed shared memory (DSM) agent is an owner of a data segment;

including a transaction journal in each node for storing committed transactions generated by users on that node;

scanning a list to identify the location of the latest contents of a data segment prior to the transaction using a rollback procedure;

for each modified data segment, identifying the location of the latest contents prior to the transaction using the rollback procedure by;

if the data segment was marked as modified in the cache at the time it was inserted into the list, then the latest contents of this data segment appears only in the journal,

otherwise, if the data segment was not marked as modified in the cache at the time the data segment was inserted into the list, then the latest contents of this data segment appears in its final location in the shared storage; and

recording a type of each data segment in the list during insertion of the data segment into the list, wherein;

all data segments in the list whose latest contents appear in a final location are discarded from the cache,

for all the other data segments in the list, latest contents for the all the other data segments are restored from the journal into the cache by scanning the local transaction journal from ending to beginning, and first occurrences of these data segments in the local transaction journal are considered, and a modification indication of the data segments is set as true, and exclusive permissions on all the data segments involved in a cancellation transaction are released.

2. The method of claim 1, wherein data segments, which are assigned to be checkpointed by the one of the plurality of file system agents, are a subset of the data segments appearing in the transaction journal associated with the one of the plurality of file system agents.

3. The method of claim 2, wherein the subset of the data segments appearing in the transaction journal associated with the one of the plurality of file system agents are the modified data segment and the committed data segment whose current owner is the one of the plurality of file system agents.

4. The method of claim 3, further including performing the checkpoint on the data segments by the one of the plurality of file system agents during one of a normal operation and upon a failure and recovery by another one of the one of the plurality of file system agents.

5. The method of claim 3, further including performing the checkpoint by the one of the plurality of file system agents only on the data segments owned by the one of the plurality of file system agents.

6. The method of claim 1, further including ensuring most recent contents of the data segment are written to a final storage location whenever the associated DSM agent owing the data segment transfers ownership of the data segment.

7. The method of claim 6, further including clearing a modification indication associated with the data segment in a cache memory, wherein ownership of the data segment is transferred only after the modification indication is cleared.

8. The method of claim 1, further including writing the data segment to a final storage location if the one of the plurality of file system agents, currently owning the modified data segment, has to remove the data segment from an associated cache slot.

9. A system for performing concurrent transactional checkpoints from a plurality of file system agents residing on different nodes within in a clustered file system (CFS), the system comprising:

a cluster of nodes,

the CFS including the cluster of nodes forming a computer cluster,

a plurality of distributed shared memory (DSM) agents, wherein each one of the plurality of DSM agents is associated with a node of the cluster of nodes,

the plurality of file system agents, wherein each one of the plurality of file system agents is in communication with each one of the plurality of DSM agents,

a plurality of storage devices in communication with the CFS,

a cache associated with the node,

a transaction journal associated with the CFS, and

a processor device having a memory coupled to the processor device for controlling the CFS, wherein the processor device is assigned to the node and the node is in communication with the plurality of storage devices, wherein the processor device;

assigns to one of the plurality of file system agents responsibility to checkpoint a modified data segment and a committed data segment to a final storage location, wherein the one of the plurality of file system agents assigned is the one of the plurality of file system agents whose associated distributed shared memory (DSM) agent is an owner of a data segment,

includes the transaction journal in each node for storing committed transactions generated by users on that node,

includes a list to identify the location of the latest contents of a data segment prior to the transaction using a rollback procedure,

for each modified data segment, identifying the location of the latest contents prior to the transaction using the rollback procedure by;

if the data segment was marked as modified in the cache at the time it was inserted into the list, then the latest contents of this data segment appears only in the journal,

otherwise, if the data segment was not marked as modified in the cache at the time the data segment was inserted into the list, then the latest contents of this data segment appears in its final location in the shared storage, and

records a type of each data segment in the list during insertion of the data segment into the list, wherein;

all data segments in the list whose latest contents appear in a final location are discarded from the cache,

for all the other data segments in the list, latest contents for the all the other data segments are restored from the journal into the cache by scanning the local transaction journal from ending to beginning, and first occurrences of these data segments in the local transaction journal are considered, and a modification indication of the data segments is set as true, and

exclusive permissions on all the data segments involved in a cancellation transaction are released.

10. The system of claim 9, wherein data segments, which are assigned to be checkpointed by the one of the plurality of file system agents, are a subset of the data segments appearing in the transaction journal associated with the one of the plurality of file system agents.

11. The system of claim 10, wherein the subset of the data segments appearing in the transaction journal associated with the one of the plurality of file system agents are the modified data segment and the committed data segment whose current owner is the one of the plurality of file system agents.

12. The system of claim 11, wherein the processor device performs the checkpoint on the data segments by the one of the plurality of file system agents during one of a normal operation and upon a failure and recovery by another one of the one of the plurality of file system agents.

13. The system of claim 11, wherein the processor device performs the checkpoint by the one of the plurality of file system agents only on the data segments owned by the one of the plurality of file system agents.

14. The system of claim 9, wherein the processor device ensures most recent contents of the data segment are written to a final storage location whenever the associated DSM agent owing the data segment transfers ownership of the data segment.

15. The system of claim 14, wherein the processor device clears a modification indication associated with the data segment in a cache memory, wherein ownership of the data segment is transferred only after the modification indication is cleared.

16. The system of claim 9, wherein the processor device writes the data segment to a final storage location if the one of the plurality of file system agents, currently owning the modified data segment, has to remove the data segment from an associated cache slot.

17. A computer program product for performing concurrent transactional checkpoints from a plurality of file system agents residing on different nodes within in a clustered file system (CFS) using a processor device, the computer program product comprising a non-transitory computer-readable storage medium having computer-readable program code portions stored therein, the computer-readable program code portions comprising:

a first executable portion that assigns to one of the plurality of file system agents responsibility to checkpoint a modified data segment and a committed data segment to a final storage location, wherein the one of the plurality of file system agents assigned is the one of the plurality of file system agents whose associated distributed shared memory (DSM) agent is an owner of a data segment;

a second executable portion that includes a transaction journal in each node for storing committed transactions generated by users on that node;

a third executable portion that includes a list to identify the location of the latest contents of a data segment prior to the transaction using a rollback procedure;

for each modified data segment, a fourth executable portion that identifies the location of the latest contents prior to the transaction using the rollback procedure by;

if the data segment was marked as modified in the cache at the time it was inserted into the list, then the latest contents of this data segment appears only in the journal,

otherwise, if the data segment was not marked as modified in the cache at the time the data segment was inserted into the list, then the latest contents of this data segment appears in its final location in the shared storage; and

a fifth executable portion that records a type of each data segment in the list during insertion of the data segment into the list, where;

all data segments in the list whose latest contents appear in a final location are discarded from the cache,

for all the other data segments in the list, latest contents for the all the other data segments are restored from the journal into the cache by scanning the local transaction journal from ending to beginning, and first occurrences of these data segments in the local transaction journal are considered, and a modification indication of the data segments is set as true, and

exclusive permissions on all the data segments involved in a cancellation transaction are released.

18. The computer program product of claim 17, wherein data segments, which are assigned to be checkpointed by the one of the plurality of file system agents, are a subset of the data segments appearing in the transaction journal associated with the one of the plurality of file system agents.

19. The computer program product of claim 18, wherein the subset of the data segments appearing in the transaction journal associated with the one of the plurality of file system agents are the modified data segment and the committed data segment whose current owner is the one of the plurality of file system agents.

20. The computer program product of claim 19, further including a sixth executable portion that performs the checkpoint on the data segments by the one of the plurality of file system agents during one of a normal operation and upon a failure and recovery by another one of the one of the plurality of file system agents.

21. The computer program product of claim 19, further including a sixth executable portion that performs the checkpoint by the one of the plurality of file system agents only on the data segments owned by the one of the plurality of file system agents.

22. The computer program product of claim 17, further including a sixth executable portion that ensures most recent contents of the data segment are written to a final storage location whenever the associated DSM agent owing the data segment transfers ownership of the data segment.

23. The computer program product of claim 22, further including a seventh executable portion that clears a modification indication associated with the data segment in a cache memory, wherein ownership of the data segment is transferred only after the modification indication is cleared.

24. The computer program product of claim 17, further including a sixth executable portion that writes the data segment to a final storage location if the one of the plurality of file system agents, currently owning the modified data segment, has to remove the data segment from an associated cache slot.