Data migration between multiple tiers in a storage system using age and frequency statistics
View Patent ↗Technologies are described for implementing a migration mechanism in a storage system containing multiple tiers of storage with each tier having different cost and performance parameters. Access statistics can be collected for each territory, or storage entity, within the storage system. Data that is accessed more frequently can be migrated toward higher performance storage tiers while data that is accessed less frequently can be migrated towards lower performance storage tiers. The placement of data may be governed first by the promotion of territories with higher access frequency to higher tiers. Secondly, data migration may be governed by demoting territories to lower tiers to create room for the promotion of more eligible territories from the next lower tier. In instances where space is not available on the next lower tier, further demotion may take place to an even lower tier in order to make space for the first demotion.
1. A method for migrating data between tiers of a storage system, the method comprising:
subdividing a storage capacity of the storage system into a set of territories;
partitioning the set of territories into tiers;
imposing an ordering on the tiers;
maintaining an information lifecycle management metric for each territory, wherein maintaining an information lifecycle management metric comprises combining an age statistic and a frequency statistic into a single metric, wherein respective contributions of the age statistic and the frequency statistic to the single metric are mutually exclusive, and wherein the single metric is a signed value to indicate whether the single metric represents the age statistic or the frequency statistic, respectively;
updating the information lifecycle management metric once per update period, wherein each update period has a consistent amount of time;
promoting a territory that is more eligible into a highest possible tier in response to storage capacity being available in a higher tier;
demoting a territory that is less eligible into a lower tier in response to a more eligible territory existing in the lower tier and no storage capacity being available for promotion of the more eligible territory; and
demoting a least eligible territory to a next lowest tier in response to a less eligible territory existing in a next highest tier and no storage capacity being available for demotion of the less eligible territory.
2. The method of claim 1 , wherein imposing an ordering on the tiers comprises ordering each one of the tiers higher in response to a respective tier having higher performance characteristics.
3. The method of claim 1 , further comprising providing a periodic timer for scheduling of data migration.
4. The method of claim 1 , further comprising providing a periodic timer for updating information lifecycle management metrics, wherein the periodic timer establishes the update period.
5. The method of claim 4 , further comprising incrementing an age statistic associated with at least one territory in response to the periodic timer for updating information lifecycle management metrics.
6. The method of claim 5 , wherein the age statistic associated with the at least one territory is incremented in response to the at least one territory not being accessed during the update period.
7. The method of claim 4 , further comprising incrementing a frequency statistic associated with at least one territory during the update period.
8. The method of claim 7 , wherein the frequency statistic associated with the at least one territory is incremented in response to the at least one territory being accessed during the update period.
9. The method of claim 1 , further comprising sorting the territories within each tier in order of the information lifecycle management metric associated with each respective territory.
10. A non-transitory computer storage medium having computer-executable instructions stored thereon which, when executed by a computer system, cause the computer system to:
subdivide a storage capacity of a storage system into a set of territories;
partition the set of territories into ordered tiers;
maintain an information lifecycle management metric for each territory, wherein maintaining an information lifecycle management metric comprises combining an age statistic and a frequency statistic into a single metric, wherein respective contributions of the age statistic and the frequency statistic to the single metric are mutually exclusive, and wherein the single metric is a signed value to indicate whether the single metric represents the age statistic or the frequency statistic, respectively;
update the information lifecycle management metric once per update period, wherein each update period has a consistent amount of time;
promote a territory that is more eligible into a highest possible tier in response to storage capacity being available in a higher tier;
demote a territory that is less eligible into a lower tier in response to a more eligible territory existing in the lower tier and no storage capacity being available for promotion of the more eligible territory; and
demote a least eligible territory to a next lowest tier in response to a less eligible territory existing in a next highest tier and no storage capacity being available for demotion of the less eligible territory.
11. The non-transitory computer storage medium of claim 10 , further causing the computer system to impose an ordering on the tiers by ordering each one of the tiers higher in response to a respective tier having higher performance characteristics.
12. The non-transitory computer storage medium of claim 10 , further causing the computer system to provide a periodic timer for scheduling of data migration.
13. The non-transitory computer storage medium of claim 10 , further causing the computer system to provide a periodic timer for updating information lifecycle management metrics, wherein the periodic timer establishes the update period.
14. The non-transitory computer storage medium of claim 13 , further causing the computer system to increment an age statistic associated with at least one territory when a frequency statistic associated with the at least one territory equals zero in response to the periodic timer for updating information lifecycle management metrics.
15. The non-transitory computer storage medium of claim 14 , wherein the age statistic associated with the at least one territory is incremented in response to the at least one territory not being accessed during the update period.
16. The non-transitory computer storage medium of claim 10 , further causing the computer system to sort the territories within each tier in order of information lifecycle management metrics associated with each respective territory.
17. A data storage system comprising:
a storage subsystem;
a plurality of physical storage devices associated with the storage subsystem; and
one or more data migration modules operable to cause the storage subsystem to:
subdivide a storage capacity of the storage system into a set of territories;
partition the set of territories into tiers;
maintain an information lifecycle management metric for each territory, wherein maintaining an information lifecycle management metric comprises combining an age statistic and a frequency statistic into a single metric, wherein respective contributions of the age statistic and the frequency statistic to the single metric are mutually exclusive, and wherein the single metric is a signed value to indicate whether the single metric represents the age statistic or the frequency statistic, respectively;
update the information lifecycle management metric once per update period, wherein each update period has a consistent amount of time;
promote a territory that is more eligible into a highest possible tier in response to storage capacity being available in a higher tier;
demote a territory that is less eligible into a lower tier in response to a more eligible territory existing in the lower tier and no storage capacity being available for promotion of the more eligible territory; and
demote a least eligible territory to a next lowest tier in response to a less eligible territory existing in a next highest tier and no storage capacity being available for demotion of the less eligible territory.
18. The data storage system of claim 17 , wherein the data migration modules are further operable to provide a periodic timer for scheduling of data migration.
19. The data storage system of claim 17 , wherein the data migration modules are further operable to provide a periodic timer for updating information lifecycle management metrics, wherein the periodic timer establishes the update period.
20. The data storage system of claim 19 , wherein the data migration modules are further operable to increment an age statistic associated with at least one territory in response to the periodic timer for updating information lifecycle management metrics.