Method and system for implementing a maintenance service for managing I/O and storage for a virtualization environment

• US 9,354,912 B1
• Filed: 07/22/2013
• Issued: 05/31/2016
• Est. Priority Date: 08/10/2011
• Status: Active Grant

• Current Assignee: Nutanix Incorporated
• Original Assignee: Nutanix Incorporated
• Inventors: Aron, Mohit, Chun, Brent, Bayapuneni, Chaitanya Venkata Krishna
• Primary Examiner(s): Kessler, Gregory A
• Application Number:
  US13/947,668
• Time in Patent Office: 1,044 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06F 2009/45579 I/O management, e.g. providing access to device drivers or storage
• CPC Class Codes: G06F 9/4555 Para-virtualisation, i.e. guest operating system has to be modified
• CPC Class Codes: G06F 9/45558 Hypervisor-specific management and integration aspects

Abstract

Disclosed is an approach for overseeing the overall state of a virtual storage system, and for taking actions as necessary to manage the health and efficient performance of that system. A curator entity is provided on a distributed basis to manage and perform these functions. A master curator manages this workload that is performed by multiple slave curators. MapReduce operations are performed to implement the curator workload. Periodically, the master curator coordinates a full scan of all the metadata in the system using three MapReduce operations. Partial scans may also be run using a single MapReduce operation.

Claims

1. A method for managing a virtual storage system implemented with a processor, comprising:

receiving information regarding metadata maintained by a plurality of nodes in the virtual storage system, wherein the plurality of nodes comprise respective distributed maintenance components;

using the distributed maintenance components to manage the metadata for the virtual storage system, in which the metadata relate to storage of data within the virtual storage system;

checking a state of the virtual storage system to identify inconsistencies within the metadata, wherein the checking is performed in parallel by the distributed maintenance components to scan the metadata; and

maintaining the state of the virtual storage system based at least in part in results of checking the state of the virtual storage system, wherein maintaining the state of the virtual storage system is performed in parallel by the distributed maintenance components to fix the inconsistencies within the metadata distributed in the plurality of nodes.

2. The method of claim 1, wherein checking the state of the virtual storage system comprises performing a full scan of the metadata.

3. The method of claim 2 in which the full scan comprises a scan of all metadata tables.

4. The method of claim 1 in which a mapreduce operation is performed to check the state of the virtual storage system.

5. The method of claim 4 in which the mapreduce operation distributes work among the distributed maintenance components.

6. The method of claim 4 in which the mapreduce operation comprises a single mapreduce directed to an extent group ID map.

7. The method of claim 1 in which the distributed maintenance components comprise a master and multiple slaves, in which the master coordinates maintenance workloads to be run by the slaves.

8. The method of claim 1 in which certain tasks are scheduled to run as background tasks.

9. The method of claim 8 in which the background tasks are resource-intensive tasks.

10. The method of claim 1 in which a distributed maintenance component from the distributed maintenance components checks a logical timestamp for an entry in the metadata before processing the entry, wherein the entry is not processed if the logical timestamp indicates a possibility that the entry is currently being operated upon by another of a distributed maintenance component from the distributed maintenance components.

11. The method of claim 10 in which the logical timestamp is a vector clock that is capable of being associated with multiple logical timestamps for plurality of nodes.

12. The method of claim 10 in which the logical timestamp comprises an incarnation value and a component identifier.

13. The method of claim 10 in which the logical timestamp comprises an operation identifier.

14. A computer program product embodied on a non-transitory computer-usable medium, the non-transitory computer-usable medium having stored thereon a sequence of instructions which, when executed by a processor causes the processor to execute a method for managing a virtual storage system, the method comprising:

receiving information regarding metadata maintained by a plurality of nodes in the virtual storage system, wherein the plurality of nodes comprise respective distributed maintenance components;

using the distributed maintenance components to manage the metadata for the virtual storage system, in which the metadata relate to storage of data within the virtual storage system;

checking a state of the virtual storage system to identify inconsistencies within the metadata, wherein the checking is performed in parallel by the distributed maintenance components to scan the metadata; and

maintaining the state of the virtual storage system based at least in part in results of checking the state of the virtual storage system, wherein maintaining the state of the virtual storage system is performed in parallel by the distributed maintenance components to fix the inconsistencies within the metadata distributed in the plurality of nodes.

15. The computer program product of claim 14, wherein checking the state of the virtual storage system comprises performing a full scan of the metadata.

16. The computer program product of claim 15 in which the full scan comprises a scan of all metadata tables.

17. The computer program product of claim 14 in which a mapreduce operation is performed to check the state of the virtual storage system.

18. The computer program product of claim 17 in which the mapreduce operation distributes work among the distributed maintenance components.

19. The computer program product of claim 17 in which the mapreduce operation comprises a single mapreduce directed to an extent group ID map.

20. The computer program product of claim 14 in which the distributed maintenance components comprise a master and multiple slaves, in which the master coordinates maintenance workloads to be run by the slaves.

21. The computer program product of claim 14 in which certain tasks are scheduled to run as background tasks.

22. The computer program product of claim 21 in which the background tasks are resource-intensive tasks.

23. The computer program product of claim 14 in which a distributed maintenance component from the distributed maintenance components checks a logical timestamp for an entry in the metadata before processing the entry, wherein the entry is not processed if the logical timestamp indicates a possibility that the entry is currently being operated upon by another of a distributed maintenance component from the distributed maintenance components.

24. The computer program product of claim 23 in which the logical timestamp is a vector clock that is capable of being associated with multiple logical timestamps for plurality of nodes.

25. The computer program product of claim 23 in which the logical timestamp comprises an incarnation value and a component identifier.

26. The computer program product of claim 23 in which the logical timestamp comprises an operation identifier.

27. A system for managing a virtual storage system, comprising:

a processor;

a memory for holding programmable code; and

wherein the programmable code includes instructions for

receiving information regarding metadata maintained by a plurality of nodes in the virtual storage system, wherein the plurality of nodes comprise respective distributed maintenance components, using the distributed maintenance components to manage the metadata for the virtual storage system, in which the metadata relate to storage of data within the virtual storage system, checking a state of the virtual storage system to identify inconsistencies within the metadata, wherein the checking is performed in parallel by the distributed maintenance components to scan the metadata; and

maintaining the state of the virtual storage system based at least in part in results of checking the state of the virtual storage system, wherein maintaining the state of the virtual storage system is performed in parallel by the distributed maintenance components to fix the inconsistencies within the metadata distributed in the plurality of nodes.

28. The system of claim 27 in which the state of the virtual storage system is checked by performing a full scan of the metadata.

29. The system of claim 28 in which the full scan comprises a scan of all metadata tables.

30. The system of claim 27 in which a mapreduce operation is performed to check the state of the virtual storage system.

31. The system of claim 30 in which the mapreduce operation distributes work among the distributed maintenance components.

32. The system of claim 30 in which the mapreduce operation comprises a single mapreduce directed to an extent group ID map.

33. The system of claim 27 in which the distributed maintenance components comprise a master and multiple slaves, in which the master coordinates maintenance workloads to be run by the slaves.

34. The system of claim 27 in which certain tasks are scheduled to run as background tasks.

35. The system of claim 34 in which the background tasks are resource-intensive tasks.

36. The system of claim 27 in which a distributed maintenance component from the distributed maintenance components checks a logical timestamp for an entry in the metadata before processing the entry, wherein the entry is not processed if the logical timestamp indicates a possibility that the entry is currently being operated upon by another of a distributed maintenance components from the distributed maintenance components.

37. The system of claim 36 in which the logical timestamp is a vector clock that is capable of being associated with multiple logical timestamps for plurality of nodes.

38. The system of claim 36 in which the logical timestamp comprises an incarnation value and a component identifier.

39. The system of claim 36 in which the logical timestamp comprises an operation identifier.

40. The system of claim 27 in which a master maintenance component comprises a master job scheduler for assigning workloads to performed on a scheduled or ad hoc basis.

41. The system of claim 40 in which the master maintenance component further comprises a map reduce job tracker for coordinating map reduce workloads across multiple slave maintenance components.

42. The system of claim 27 in which the distributed maintenance components comprise a map reduce job tracker for performing map reduce workloads.

43. The system of claim 27 in which one or more remote procedure calls are performed to communicate workloads between the distributed maintenance components.