ASYNCHRONOUS FILE OPERATIONS IN A SCALABLE MULTI-NODE FILE SYSTEM CACHE FOR A REMOTE CLUSTER FILE SYSTEM

• US 20110145499A1
• Filed: 12/16/2009
• Published: 06/16/2011
• Est. Priority Date: 12/16/2009
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Hildebrand, Dean, Schmuck, Frank B., Tewari, Renu, Haskin, Roger L., Ananthanarayanan, Rajagopol, Naik, Manoj P., Eshel, Marc M.
• Granted Patent: US 8,495,250 B2
• Time in Patent Office: Days
• US Class Current: 711/118
• CPC Class Codes: G06F 12/0866 for peripheral storage systems, e.g. disk cache
• CPC Class Codes: G06F 16/172 Caching, prefetching or hoarding of files
• CPC Class Codes: G06F 16/1858 Parallel file systems, i.e. file systems supporting multiple processors
• CPC Class Codes: G06F 2212/263 Network storage, e.g. SAN or NAS

Abstract

Asynchronous file operations in a scalable multi-node file system cache for a remote cluster file system, is provided. One implementation involves maintaining a scalable multi-node file system cache in a local cluster file system, and caching local file data in the cache by fetching file data on demand from the remote cluster file system into the cache over the network. The local file data corresponds to file data in the remote cluster file system. Local file information is asynchronously committed from the cache to the remote cluster file system over the network.

Claims

1. A method for facilitating access to data in a wide area network, comprising:

maintaining a scalable multi-node file system cache in a local cluster file system including multiple computing nodes;

caching local file data in the cache by fetching file data on demand from a remote cluster file system into the cache over the network, wherein the local file data and metadata corresponds to the file data and metadata in the remote cluster file system, respectively; and

asynchronously committing local file information from the cache to the remote cluster file system over the network.

2. The method of claim 1 further comprising:

asynchronously committing operations involving data and metadata, with one or more computing nodes of the file system cache; and

coordinating asynchronously committing operations to maintain consistency.

3. The method of claim 1 further comprising:

coordinating and load-balancing asynchronous committing of local file information from the cache to the remote cluster file system across a number of network nodes.

4. The method of claim 1 further comprising:

asynchronously committing local file information from the cache to the remote cluster file system further by;

an application node of the local file system receiving a request for writing file data;

the application node writing the file data to the cache and signaling a gateway node to write the file data to the remote cluster file system;

thereafter, the gateway node asynchronously transferring the file data from the cache to the remote cluster file system over the network.

5. The method of claim 4 further comprising:

performing asynchronous update support operations out-of-order by coalescing data write operations and filtering out inverse operations.

6. The method of claim 4 further comprising:

for dependent metadata operations;

each gateway node maintaining a queue of asynchronous data or metadata object update requests by one or more application nodes;

detecting potential interdependency between objects of a set of update requests;

upon detecting that the objects are not interdependent, asynchronously servicing the update requests in any order; and

upon detecting that the data objects are interdependent, asynchronously servicing the update requests in the order in which they were executed on the application nodes.

7. The method of claim 1 further comprising:

upon detecting that a requested file data is not in the cache, fetching file data on demand from the remote cluster file system into the cache over the network.

8. The method of claim 7, wherein caching local file data in the cache further includes:

an application node of the local cached file system receiving a request for file access;

the application node checking the cache for the requested file;

upon detecting a cache hit, the application node accessing the file data in the cache;

upon detecting a cache miss, fetching file data on demand from the remote cluster file system into the local cached file system over the network, wherein fetching file data on demand includes the application node signaling a gateway node to fetch the file data and the gateway node fetching the file data from the remote cluster file system into the cache over the network for access by the application node.

9. The method of claim 8, wherein:

fetching the file data from the remote cluster file system further includes;

fetching the file data from the remote cluster file system into the cache over a wide area network using a parallel network file system data transfer;

the local cluster filer system and the remote cluster file system comprise General Parallel File Systems;

the cache comprises a storage subsystem of the local cluster, wherein the cache is visible to any file system client node as a Portable Operating System Interface compliant file system, allowing a file system client to browse the cache and access the cached remote file data as in a local file system.

10. A system for facilitating access to data in a wide area network, comprising:

a cache subsystem comprising in a cache for a local cluster file system including multiple computing nodes;

a caching layer function configured for providing scalable multi-node file system caching in the local cluster file system by;

caching local file data in the cache by fetching file data on demand from a remote cluster file system into the cache over the network, wherein the local file data corresponds to file data in the remote cluster file system; and

asynchronously committing local file information from the cache to the remote cluster file system over the network.

11. The system of claim 10, wherein the caching layer function is further configured for asynchronously committing operations involving data and metadata, wherein one or more computing nodes of the cache coordinate asynchronously committing operations to maintain consistency.

12. The system of claim 10, wherein the cache layer function includes an application node and a gateway node in the local cached file system, the application node configured for receiving a request for writing file data, writing the file data to the cache and signaling a gateway node to write the file data to the remote cluster file system, such that the gateway node asynchronously transfers said file data from the cache to the remote cluster file system over the network.

13. The system of claim 12, wherein the each gateway node is further configured such that for dependent metadata operations, each gateway node:

maintains a queue of asynchronous data or metadata object update requests by one or more application nodes;

detects potential interdependency between objects of a set of update requests;

upon detecting that the objects are not interdependent, asynchronously services the update requests in any order; and

upon detecting that the data objects are interdependent, asynchronously services the update requests in the order in which they were executed on the application nodes.

14. The system of claim 10, wherein:

the caching layer function is further configured such that, upon detecting that a requested file data is not in the cache, the caching layer function fetches file data on demand from the remote cluster file system into the cache over the network;

the caching layer function includes a gateway node, such that upon detecting a cache miss the application node invokes the gateway node for fetching file data on demand from the remote cluster file system into the cache over the network, wherein fetching file data on demand includes the application node signaling the gateway node to fetch the file data and the gateway node fetching the file data from the remote cluster file system into the cache over the network for access by the application node;

the gateway node is further configured for fetching the file data from the remote cluster file system into the cache over a wide area network using a parallel network file system data transfer; and

the local cluster filer system and the remote cluster file system comprise General Parallel File Systems, and the cache is visible to any file system client node as a Portable Operating System Interface compliant file system, allowing a file system client to browse the cache and access the cached remote file data as in a local file system.

15. A computer program product for facilitating access to data, the computer program product comprising:

a computer readable storage medium having computer usable program code embodied therewith, the computer usable program code comprising;

computer usable program code configured to provide a cache subsystem for maintaining data in a cache for a local cluster file system;

computer usable program code configured to provide a caching layer function for caching local file data in the cache by fetching file data on demand from a remote cluster file system into the cache over the network, wherein the local file data corresponds to file data in the remote cluster file system; and

asynchronously commit local file information from the cache to the remote cluster file system over the network.

16. The computer program product of claim 15 further comprising:

computer usable program code configured to asynchronously commit local file information from the cache to the remote cluster file system by asynchronously committing operations involving data and metadata, with one or more computing nodes of the cache coordinating asynchronously committing operations to maintain consistency.

17. The computer program product of claim 15 further comprising:

computer usable program code configured for an application node of the local cached file system to receive a request for writing file data, write the file data to the cache, and signal a gateway node to write the file data to the remote cluster file system, the program instructions further causing the gateway node to asynchronously transfer said file data from the cache to the remote cluster file system over the network.

18. The computer program product of claim 15 further comprising:

computer usable program code configured for dependent metadata operations, wherein each gateway node maintains a queue of asynchronous data or metadata object update requests by one or more application nodes, detects potential interdependency between objects of a set of update requests, upon detecting that the objects are not interdependent, asynchronously services the update requests in any order, otherwise upon detecting that the data objects are interdependent, asynchronously services the update requests in the order in which they were executed on the application nodes.

19. The computer program product of claim 18 further comprising:

computer usable program code configured to detect if a requested file data is not in the cache, and fetching file data on demand from the remote cluster file system into the cache over the network; and

transfer local file data from the cache to the remote cluster file system over the network.

20. The computer program product of claim 19 further comprising:

computer usable program code configured for an application node to receive a request for file access, check the cache for the requested file, and upon detecting a cache hit, access the file data in the cache;

computer usable program code configured for an application node to detect a cache miss and invoke a gateway node for fetching file data on demand from the remote cluster file system into the cache over the network, wherein fetching file data on demand includes the application node signaling the gateway node to fetch the file data and the gateway node fetching the file data from the remote cluster file system into the cache over the network for access by the application node; and

computer usable program code configured for the gateway node to fetch the file data from the remote cluster file system into the cache over a wide area network using a parallel network file system data transfer;

wherein the local cluster filer system and the remote cluster file system comprise General Parallel File Systems, and the cache is visible to any file system client node as a Portable Operating System Interface compliant file system, allowing a file system client to browse the cache and access the cached remote file data as in a local file system.