Optimizing access time of files stored on storages

• US 8,533,183 B2
• Filed: 03/10/2009
• Issued: 09/10/2013
• Est. Priority Date: 03/10/2009
• Status: Active Grant

• Current Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: Hokanson, Mark K.
• Primary Examiner(s): Wilson, Kimberly
• Assistant Examiner(s): HOANG, HAU HAI
• Application Number:
  US13/203,978
• Publication Number:
  US 20110320436A1
• Time in Patent Office: 1,645 Days
• Field of Search: 711/117, 711/165, 707/698, 707/705
• US Class Current: 707/719
• CPC Class Codes: G06F 12/0246 in block erasable memory, e.g. flash memory
• CPC Class Codes: G06F 12/122 of the least frequently used [LFU] type, e.g. with individual count value
• CPC Class Codes: G06F 3/061 Improving I/O performance
• CPC Class Codes: G06F 3/0647 Migration mechanisms
• CPC Class Codes: G06F 3/0685 Hybrid storage combining heterogeneous device types, e.g. hierarchical storage, hybrid arrays

Abstract

When each file of a number of files is accessed, at least a number of times each file has been accessed is kept track of. Each file is stored on a storage of a number of storages. Periodically, at least one file is moved among the number of storages, based at least on the number of times each file has been accessed. As such, the at least one file is moved from being stored on a first storage to being stored on a second storage, to optimize subsequent access time of the at least one file. The storages are physically distinct storage devices. At least one of the storage devices has different storage characteristics as compared to one or more other of the storage devices.

Claims

1. A method comprising:

when each file of a plurality of files is accessed, keeping track of at least a number of times each file has been accessed, where each file is stored on a storage of a plurality of storages; and

,

periodically moving at least one file of the plurality of files among the plurality of storages in accordance with an optimization strategy to optimize subsequent access time of the at least one file,

wherein the plurality of storages comprises a plurality of physically distinct storage devices, at least one of the storage devices having different storage characteristics as compared to one or more other of the storage devices,

wherein the optimization strategy is one in which the at least one file that are periodically moved are selected based on ratios of numbers of accesses thereof to file sizes thereof and comprise a first file having a first ratio of a number of accesses of the first file to a file size of the first file and a second file having a second ratio of a number of accesses of the second file to a file size of the second file, the first ratio greater than the second ratio, the first file moved to a third first storage of the plurality of storages having a first access time and a first capacity, the second file moved to a second storage of the plurality of storages having a second access time and a second capacity, the first access time greater than the second access time, the first capacity less than the second capacity.

2. The method of claim 1, further comprising maintaining the storage characteristics of each storage of the plurality of storages.

3. The method of claim 2, wherein maintaining the storage characteristics of each storage of the plurality of storages comprises, for each storage, storing one or more of:

a sequential read speed of the storage;

a sequential write speed of the storage;

a random-access read speed of the storage;

a random-access write speed of the storage; and

,

a size of the storage.

4. The method of claim 2, wherein periodically moving the plurality of files among the plurality of storages is further based on the storage characteristics of each storage.

5. The method of claim 1, wherein keeping track of at least the number of times each file has been accessed further comprises keeping track of a size of each file that has been accessed, and

wherein periodically moving the plurality of files among the plurality of storages is further based on the size of each file that has been accessed.

6. The method of claim 1, wherein keeping track of at least the number of times each file has been accessed comprises relying upon an operating system tracking functionality that keeps track of at least the number of times each file has been accessed.

7. The method of claim 1, wherein keeping track of at least the number of times each file has been accessed comprises:

detecting that a given file of the plurality of files has been accessed; and

,

responsive to detecting that the given file has been accessed, incrementing the number of times that the given file has been accessed.

8. The method of claim 1, wherein keeping track of at least the number of times each file has been accessed comprises keeping track of a number of times each file has been read from and keeping track of a number of times each file has been written to, and

wherein periodically moving the plurality of files among the plurality of storages is further based on the number of times each file has been read from and the number of times each file has been written to.

9. The method of claim 1, wherein keeping track of at least the number of times each file has been accessed comprises further keeping track of caching characteristics regarding each file, and

wherein periodically moving the plurality of files among the plurality of storages is further based on the caching characteristics regarding each file.

10. The method of claim 9, wherein the caching characteristics regarding each file comprises one or more of:

whether each file has been cached;

whether each file is currently being cached;

a number of times each file has been cached; and

,

a length of time each file has been cached.

11. The method of claim 1, wherein periodically moving the at least one file of the plurality of files among the plurality of storages comprises moving a sub-plurality of the files that have been accessed a greatest number of times to one or more of the storage devices having fastest access speeds.

12. The method of claim 1, wherein periodically moving the at least one file of the plurality of files among the plurality of storages comprises moving a sub-plurality of the files that have been accessed a greatest number of times to one or more of the storage devices having fastest access speeds, but taking into account sizes of the sub-plurality of the files and sizes of the storage devices so that no one file takes up all storage space on a given storage of the plurality of storages.

13. The method of claim 1, wherein periodically moving the at least one file of the plurality of files among the plurality of storages comprises one or more of:

periodically moving the at least one file of the plurality of files among the plurality of storages at regularly scheduled times in accordance with a predetermined schedule; and

,

periodically moving the at least one file of the plurality of files among the plurality of storages responsive to manual user initiation.

14. A non-transitory computer-readable medium having a computer program stored thereon for execution by one or more processors, such that execution of the computer program by the processors from the computer-readable medium causes the computer program to perform a method comprising:

when each file of a plurality of files is accessed, keeping track of at least a number of times each file has been accessed, where each file is stored on a storage of a plurality of storages; and

,

periodically moving at least one file of the plurality of files among the plurality of storages in accordance with an optimization strategy to optimize subsequent access time of the at least one file,

wherein the plurality of storages comprises a plurality of physically distinct storage devices, at least one of the storage devices having different storage characteristics as compared to one or more other of the storage devices,

wherein the optimization strategy is one in which the at least one file that are periodically moved are selected based on ratios of numbers of accesses thereof to file sizes thereof and comprise a first file having a first ratio of a number of accesses of the first file to a file size of the first file and a second file having a second ratio of a number of accesses of the second file to a file size of the second file, the first ratio greater than the second ratio, the first file moved to a first storage of the plurality of storages having a first access time and a first capacity, the second file moved to a second storage of the plurality of storages having a second access time and a second capacity, the first access time greater than the second access time, the first capacity less than the second capacity.

15. A computing system comprising:

a plurality of storages comprising a plurality of physically distinct storage devices, at least one of the storage devices having different storage characteristics as compared to one or more other of the storage devices, the plurality of storages storing a plurality of files that are accessed; and

,

a mechanism to periodically move at least one file of the plurality of files among the plurality of storages in accordance with an optimization strategy to optimize subsequent access time of the at least one file,

wherein the optimization strategy is one in which the at least one file that are periodically moved are selected based on ratios of numbers of accesses thereof to file sizes thereof and comprise a first file having a first ratio of a number of accesses of the first file to a file size of the first file and a second file having a second ratio of a number of accesses of the second file to a file size of the second file, the first ratio greater than the second ratio, the first file moved to a first storage of the plurality of storages having a first access time and a first capacity, the second file moved to a second storage of the plurality of storages having a second access time and a second capacity, the first access time greater than the second access time, the first capacity less than the second capacity.