Managing data in a multi-level raid storage array

• US 20030221060A1
• Filed: 05/23/2002
• Published: 11/27/2003
• Est. Priority Date: 05/23/2002
• Status: Active Grant

• Current Assignee: Hewlett Packard Enterprise Development LP (Hewlett-Packard Enterprise Company)
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: Navarro, Guillermo, Umberger, David K., Condel, Jonathan
• Granted Patent: US 6,898,667 B2
• Time in Patent Office: Days
• US Class Current: 711/114
• CPC Class Codes: G06F 3/0608 Saving storage space on storage systems
• CPC Class Codes: G06F 3/0613 in relation to throughput
• CPC Class Codes: G06F 3/0647 Migration mechanisms
• CPC Class Codes: G06F 3/0653 Monitoring storage devices or systems
• CPC Class Codes: G06F 3/0689 Disk arrays, e.g. RAID, JBOD

Abstract

A system and methods implemented within a multi-level RAID (redundant array of independent disks) storage array operate to initially write data to a lower performing RAID level within the array. In addition, data is migrated between lower and higher performing RAID levels via data migration processes that function as background processes. Benefits of the disclosed system and methods include a non-disruptive environment for servicing host I/O (input/output) requests. Array response times are significantly reduced by not allowing initial data writes to interfere with higher performing RAID levels and by migrating data between lower and higher performing RAID levels in the background when the array is less busy servicing host I/O requests.

Claims

1. A processor-readable medium in a multi-level RAID (redundant array of independent disks) storage array comprising processor-executable instructions configured for:

initiating a data migration process;

searching for migratable data;

checking an array utilization rate if migratable data is found; and

migrating the migratable data if the array utilization rate does not exceed a minimum threshold.

2. A processor-readable medium as recited in claim 1, comprising further processor-executable instructions configured for terminating the data migration process if migratable data is not found.

3. A processor-readable medium as recited in claim 2, comprising further processor-executable instructions configured for reinitiating the data migration process based on factors selected from a group of factors comprising:

time; and

events.

4. A processor-readable medium as recited in claim 1, comprising further processor-executable instructions configured for terminating the data migration process if the array utilization rate is above the minimum threshold.

5. A processor-readable medium as recited in claim 4, comprising further processor-executable instructions configured for reinitiating the data migration process based on factors selected from a group of factors comprising:

time; and

events.

6. A processor-readable medium as recited in claim 1, wherein the migrating further comprises:

promoting newer migratable data to a higher performing RAID level; and

demoting older migratable data to a lower performing RAID level.

7. A processor-readable medium as recited in claim 1, comprising further processor-executable instructions configured for initially storing newly written data in a lower performing RAID level.

8. A processor-readable medium as recited in claim 1, comprising further processor-executable instructions configured for triggering the initiating based on factors selected from a group of factors comprising:

time; and

events.

9. A processor-readable medium as recited in claim 1, wherein the searching further comprises comparing the age of data in a low performing RAID level with the age of data in a higher performing RAID level.

10. A processor-readable medium as recited in claim 1, wherein the locating further comprises comparing the age of data with a threshold age.

11. A processor-readable medium in a multi-level RAID (redundant array of independent disks) storage array comprising processor-executable instructions configured for:

periodically initiating a data promotion process;

as part of the data promotion process, determining if data blocks in a lower performing RAID level are promotion candidates;

if the lower performing RAID level contains a promotion candidate, initiating a utilization check process to monitor array utilization; and

if the array utilization reaches a threshold level, promoting promotion candidates from the lower performing RAID level to a higher performing RAID level.

12. A processor-readable medium as recited in claim 11, comprising further processor-executable instructions configured for:

periodically initiating a data demotion process;

as part of the data demotion process, determining if old data blocks in the higher performing RAID level are demotion candidates;

if the higher performing RAID level contains a demotion candidate, initiating the utilization check process to monitor array utilization; and

if the array utilization reaches the threshold level, demoting demotion candidates from the higher performing RAID level to the lower performing RAID level.

13. A processor-readable medium as recited in claim 11, comprising further processor-executable instructions configured for terminating the promotion process if the lower performing RAID level does not contain a promotion candidate.

14. A processor-readable medium as recited in claim 12, comprising further processor-executable instructions configured for terminating the demotion process if the higher performing RAID level does not contain a demotion candidate.

15. A processor-readable medium as recited in claim 11, comprising further processor-executable instructions configured for initially writing incoming data updates to the lower performing RAID level.

16. A processor-readable medium as recited in claim 11, wherein the determining if data blocks in a lower performing RAID level are promotion candidates further comprises locating most recently updated data blocks.

17. A processor-readable medium as recited in claim 11, wherein the determining if data blocks in a lower performing RAID level are promotion candidates further comprises comparing an update time for each data block with an update time threshold.

18. A processor-readable medium as recited in claim 12, wherein the determining if old data blocks in the higher performing RAID level are demotion candidates further comprises comparing an update time of the old data blocks with an update time threshold.

19. A processor-readable medium as recited in claim 12, wherein the data promotion process and the data demotion process are initiated concurrently.

20. A processor-readable medium as recited in claim 11, comprising further processor-executable instructions configured for terminating the promotion process when a higher priority process occurs.

21. A processor-readable medium as recited in claim 12, comprising further processor-executable instructions configured for terminating the demotion process when a higher priority process occurs.

22. A processor-readable medium in a multi-level RAID (redundant array of independent disks) storage array comprising processor-executable instructions configured for:

periodically initiating a data demotion process;

as part of the data demotion process, determining if old data blocks in a higher performing RAID level are demotion candidates;

if the higher performing RAID level contains a demotion candidate, initiating a utilization check process to monitor array utilization; and

if the array utilization reaches a threshold level, demoting demotion candidates from the higher performing RAID level to a lower performing RAID level.

23. A processor-readable medium in a multi-level RAID (redundant array of independent disks) storage array comprising processor-executable instructions configured for:

periodically checking age of data stored in a higher performing RAID level and a lower performing RAID level;

if a lower performing RAID level data block has been updated more recently than a higher performing RAID level data block, initiating a utilization check process to monitor array utilization; and

if the array utilization reaches a threshold level, promoting the lower performing RAID level data block to the higher performing RAID level and demoting the higher performing RAID level data block to the lower performing RAID level.

24. A method for migrating data in a multi-level RAID (redundant array of independent disks) storage array comprising:

initiating a data migration process in a multi-level RAID (redundant array of independent disks) storage array;

searching for migratable data;

checking an array utilization rate if migratable data is found; and

migrating the migratable data if the array utilization rate is at or below a minimum threshold.

25. A method of managing data in a multi-level RAID (redundant array of independent disks) storage array comprising:

periodically initiating a data promotion process;

as part of the data promotion process, determining if data blocks in a lower performing RAID level are promotion candidates;

if the lower performing RAID level contains a promotion candidate, initiating a utilization check process to monitor array utilization; and

if the array utilization reaches a threshold level, promoting promotion candidates from the lower performing RAID level to a higher performing RAID level.

26. A method as recited in claim 25, further comprising:

periodically initiating a data demotion process;

as part of the data demotion process, determining if old data blocks in the higher performing RAID level are demotion candidates;

if the higher performing RAID level contains a demotion candidate, initiating the utilization check process to monitor array utilization; and

if the array utilization reaches the threshold level, demoting demotion candidates from the higher performing RAID level to the lower performing RAID level.

27. A method of managing data in a multi-level RAID (redundant array of independent disks) storage array comprising:

periodically checking the age of data stored in a higher performing RAID level and a lower performing RAID level;

if a lower performing RAID level data block has been updated more recently than a higher performing RAID level data block, initiating a utilization check process to monitor array utilization; and

if the array utilization reaches a threshold level, promoting the lower performing RAID level data block to the higher performing RAID level and demoting the higher performing RAID level data block to the lower performing RAID level.

28. A virtual RAID (redundant array of independent disks) storage array comprising:

disk drives employing a lower performing RAID level and a higher performing RAID level;

a data migration module configured to determine migratable data and migrate the migratable data between RAID levels; and

a utilization module to monitor a utilization rate of the array and determine when to migrate the migratable data.

29. A virtual RAID array as recited in claim 28, further comprising:

time and/or event information to trigger execution of the data migration module; and

utilization threshold information to compare to the utilization rate.

30. A virtual storage array comprising:

storage components employing a lower performing storage level and a higher performing storage level;

a data migration module configured to determine migratable data and migrate the migratable data between storage levels; and

a utilization module to monitor a utilization rate of the array and determine when to migrate the migratable data.

31. A virtual storage array as recited in claim 30, further comprising:

time and event information to trigger execution of the data migration module; and

utilization threshold information to compare to the utilization rate.

32. A virtual storage array as recited in claim 30, wherein the storage components are selected from a group of components comprising:

magnetic disks;

tapes;

optical disks; and

solid state disks.

33. A virtual RAID (redundant array of independent disks) storage array comprising:

disk drives employing a lower performing RAID level and a higher performing RAID level;

a data promotion module configured to execute periodically to determine promotion candidates and to promote the promotion candidates from the lower performing RAID level to the higher performing RAID level at a time when the array is not busy; and

a utilization module configured to monitor a utilization rate of the array and to determine the time when the array is not busy.

34. A virtual RAID (redundant array of independent disks) storage array comprising:

disk drives employing a lower performing RAID level and a higher performing RAID level;

a data demotion module configured to execute periodically to determine demotion candidates and to demote the demotion candidates from the higher performing RAID level to the lower performing RAID level at a time when the array is not busy; and

a utilization module configured to monitor a utilization rate of the array and to determine the time when the array is not busy.

35. A system comprising:

a RAID (redundant array of independent disks) storage array to service data requests from a host computer, the RAID storage array having disk drives employing a lower performing RAID level and a higher performing RAID level;

a data migration module configured to execute periodically and migrate data between RAID levels; and

a utilization module configured to execute when migratable data is located and to inform the data migration module when to migrate the migratable data.

36. A multi-level RAID (redundant array of independent disks) storage array comprising:

means for initiating a data migration process;

means for searching for migratable data;

means for checking an array utilization rate if migratable data is found; and

means for migrating the migratable data if the array utilization rate does not exceed a minimum threshold.

37. A multi-level RAID storage array as recited in claim 36, further comprising means for terminating the data migration process if migratable data is not found.

38. A multi-level RAID storage array as recited in claim 37, further comprising means for reinitiating the data migration process based on factors selected from a group of factors comprising:

time; and

events.

39. A multi-level RAID storage array as recited in claim 36, further comprising means for terminating the data migration process if the array utilization rate is above the minimum threshold.

40. A multi-level RAID storage array as recited in claim 39, further comprising means for reinitiating the data migration process based on factors selected from a group of factors comprising:

time; and

events.

41. A multi-level RAID storage array as recited in claim 36, wherein the means for migrating further comprises:

means for promoting newer migratable data to a higher performing RAID level; and

means for demoting older migratable data to a lower performing RAID level.

42. A multi-level RAID storage array as recited in claim 36, further comprising means for initially storing newly written data in a lower performing RAID level.

43. A multi-level RAID storage array as recited in claim 36, further comprising means for triggering the initiating based on factors selected from a group of factors comprising:

time; and

events.

44. A multi-level RAID storage array as recited in claim 36, wherein the means for searching further comprises means for comparing the age of data in a low performing RAID level with the age of data in a higher performing RAID level.

45. A multi-level RAID storage array as recited in claim 36, wherein the means for locating further comprises means for comparing the age of data with a threshold age.