Techniques for reliability and availability assessment of data storage configurations

• US 8,880,801 B1
• Filed: 09/28/2011
• Issued: 11/04/2014
• Est. Priority Date: 09/28/2011
• Status: Active Grant

• Current Assignee: Emc IP Holding Company LLC (Dell Technologies Inc.)
• Original Assignee: EMC Corporation (Dell Technologies Inc.)
• Inventors: Aharoni, Dan, Granlund, Kevin E., Robins, Edward S., Rytenkov, Sergey, Dagan, Hagay
• Primary Examiner(s): Tran, Denise
• Application Number:
  US13/200,658
• Time in Patent Office: 1,133 Days
• Field of Search: None
• US Class Current: 711/114
• CPC Class Codes: G06F 11/1092 Rebuilding, e.g. when physically replacing a failing disk
• CPC Class Codes: G06F 12/0246 in block erasable memory, e.g. flash memory
• CPC Class Codes: G06F 2212/7209 Validity control, e.g. using flags, time stamps or sequence numbers

Abstract

Described are techniques for selecting a storage pool configuration. A set of one or more candidate storage pool configurations is determined. For each of the candidate storage pool configurations of the set, a reliability calculation is performed to determine a reliability value indicating reliability of the candidate storage pool configuration. For each of the candidate storage pool configurations of the set, an availability calculation is performed to determine an availability value indicating availability of the candidate storage pool configuration. In accordance with selection criteria, one of the candidate storage pool configurations of the set for a first storage pool is selected. The selection criteria includes the availability value and the reliability value for each of the plurality of candidate storage pool configurations.

Claims

1. A method of selecting a storage pool configuration comprising:

determining a set of one or more candidate storage pool configurations;

performing, for each of the candidate storage pool configurations of the set, a reliability calculation to determine a reliability value indicating reliability of said each candidate storage pool configuration;

performing, for each of the candidate storage pool configurations of the set, an availability calculation to determine an availability value indicating availability of said each candidate storage pool configuration; and

selecting, in accordance with selection criteria, one of the candidate storage pool configurations of the set for a first storage pool, said selection criteria including said availability value and said reliability value for each of said plurality of candidate storage pool configurations, and wherein said selecting further comprises;

determining that said reliability value for said one candidate storage pool configuration represents a level of storage pool reliability that is greater than an acceptable reliability threshold; and

determining that said availability value for said one candidate storage pool configuration represents a level of storage pool availability that is greater than an acceptable availability threshold.

2. The method of claim 1, wherein the selection criteria includes any of performance and cost.

3. The method of claim 1, wherein each of the one or more candidate storage pool configurations includes one or more RAID group configurations, each of said one or more RAID group configurations including a plurality of physical storage devices of one of a plurality of different types of storage media.

4. The method of claim 3, wherein the plurality of different types of storage media include solid state storage device media and one or more types of different rotating disk media.

5. The method of claim 1, further comprising providing a user interface that displays, for a storage pool including a set of physical drives of a particular type, different combinations of reliability and availability in accordance with different storage pool configurations for the storage pool, the different storage pool configurations being based on a RAID level, a RAID group configuration for the RAID level, and total storage capacity of the set of physical drives in the storage pool, wherein for each of said different storage pool configurations, a first visual indicator indicates availability of the storage pool in said each different storage pool configuration and a second visual indicator indicates reliability of the storage pool in said each different storage pool configuration, said first visual indicator being displayed to overlay said second visual indicator on a display area.

6. A method of selecting a storage pool configuration comprising:

determining a set of one or more candidate storage pool configurations;

performing, for each of the candidate storage pool configurations of the set, a reliability calculation to determine a reliability value indicating reliability of said each candidate storage pool configuration;

performing, for each of the candidate storage pool configurations of the set, an availability calculation to determine an availability value indicating availability of said each candidate storage pool configuration; and

selecting, in accordance with selection criteria, one of the candidate storage pool configurations of the set for a first storage pool, said selection criteria including said availability value and said reliability value for each of said plurality of candidate storage pool configurations, wherein each of the one or more candidate storage pool configurations includes one or more RAID group configurations, each of said one or more RAID group configurations including a plurality of physical storage devices of one of a plurality of different types of storage media, wherein the plurality of different types of storage media include solid state storage device media and one or more types of different rotating disk media, and wherein said one candidate storage pool configuration includes a plurality of RAID groups, and the method further comprising;

determining, for each portion of said plurality of RAID groups having a same type of storage media, a first availability value indicating a combined availability for said each portion; and

determining a storage pool availability indicating an availability for said one candidate storage pool based on one or more of said first availability values.

7. The method of claim 6, wherein a first RAID group in said each portion includes physical storage devices of a first type of storage media having a first RAID group configuration and first RAID protection level and a second RAID group in said each portion includes physical storage devices of the first type of storage media having a second RAID group configuration and a second RAID protection level, wherein said first RAID protection level differs from said second RAID protection level and/or said first RAID group configuration differs from the second RAID group configuration.

8. A method of selecting a storage pool configuration comprising:

determining a set of one or more candidate storage pool configurations;

performing, for each of the candidate storage pool configurations of the set, a reliability calculation to determine a reliability value indicating reliability of said each candidate storage pool configuration;

performing, for each of the candidate storage pool configurations of the set, an availability calculation to determine an availability value indicating availability of said each candidate storage pool configuration; and

selecting, in accordance with selection criteria, one of the candidate storage pool configurations of the set for a first storage pool, said selection criteria including said availability value and said reliability value for each of said plurality of candidate storage pool configurations, wherein each of the one or more candidate storage pool configurations includes one or more RAID group configurations, each of said one or more RAID group configurations including a plurality of physical storage devices of one of a plurality of different types of storage media, wherein the plurality of different types of storage media include solid state storage device media and one or more types of different rotating disk media, and wherein said one candidate storage pool configuration includes a plurality of RAID groups, and the method further comprising;

determining, for each portion of said plurality of RAID groups having a same type of storage media, a first reliability value indicating a combined reliability for said each portion; and

determining a storage pool reliability indicating a reliability for said one candidate storage pool based on one or more of said first reliability values.

9. A method of selecting a storage pool configuration comprising:

determining a set of one or more candidate storage pool configurations;

performing, for each of the candidate storage pool configurations of the set, a reliability calculation to determine a reliability value indicating reliability of said each candidate storage pool configuration;

performing, for each of the candidate storage pool configurations of the set, an availability calculation to determine an availability value indicating availability of said each candidate storage pool configuration; and

selecting, in accordance with selection criteria, one of the candidate storage pool configurations of the set for a first storage pool, said selection criteria including said availability value and said reliability value for each of said plurality of candidate storage pool configurations, wherein said first storage pool has a first storage pool availability and a first storage pool reliability, and wherein a second storage pool has a second storage pool availability and a second storage pool reliability, and the method further comprising;

determining, using said first storage pool availability and said second storage pool availability, a net availability indicating a combined availability for said first storage pool and said second storage pool; and

determining whether said net availability represents a level of availability that is greater than an acceptable availability threshold.

10. The method of claim 9, further comprising:

determining, using said first storage pool reliability and said second storage pool reliability, a net reliability indicating a combined reliability for said first storage pool and said second storage pool; and

determining whether said net reliability represents a level of reliability that is greater than an acceptable reliability threshold.

11. The method of claim 10, wherein if either said net reliability is not greater than the acceptable reliability threshold or said net availability is not greater than the acceptable availability threshold, determining that a combination of currently select storage pool configurations for the first storage pool and the second storage pool are unacceptable.

12. The method of claim 9, wherein if said net availability is not greater than the acceptable availability threshold, the method further includes:

partitioning said second storage pool into a plurality of smaller storage pools;

determining a small storage pool availability for each of said plurality of smaller storage pools;

determining a combined small pool availability using a plurality of small storage pool availabilities for said plurality of smaller storage pools;

determining, using said combined small pool availability and said first storage pool availability, a revised net availability indicating a combined availability for said first storage pool and said plurality of smaller storage pools; and

determining whether said revised net availability represents a level of availability that is greater than an acceptable availability threshold.

13. The method of claim 10, wherein if said net reliability is not greater than the acceptable reliability threshold, the method further includes:

selecting one of said plurality of different types of storage media of drives included in a first RAID group of any of the first storage pool and the second storage pool;

selecting a revised RAID configuration providing an increased level of protection in comparison to a current RAID configuration for the first RAID group;

updating any of said first storage pool reliability and said second storage pool reliability in accordance with said revised RAID configuration;

determining a revised net reliability using the first storage pool reliability and the second storage pool reliability; and

determining whether said revised net availability represents a level of availability that is greater than an acceptable availability threshold.

14. A method of selecting a storage pool configuration comprising:

determining a set of one or more candidate storage pool configurations;

performing, for each of the candidate storage pool configurations of the set, a reliability calculation to determine a reliability value indicating reliability of said each candidate storage pool configuration;

performing, for each of the candidate storage pool configurations of the set, an availability calculation to determine an availability value indicating availability of said each candidate storage pool configuration; and

selecting, in accordance with selection criteria, one of the candidate storage pool configurations of the set for a first storage pool, said selection criteria including said availability value and said reliability value for each of said plurality of candidate storage pool configurations, and wherein performing an availability calculation for a first of the candidate storage pool configurations of the set includes determining a first time to restore data for said first candidate pool configuration upon an occurrence of a failure of any RAID group in said first candidate storage pool configuration thereby indicating that the failing RAID group is unable to recover desired data using RAID-group internal rebuild processes, and a second time to rebuild data for a first RAID group in said first candidate storage pool configuration upon an occurrence of an internal RAID group failure of the first RAID group where the first RAID group is able to recover desired data using RAID-group internal rebuild processes for the first RAID group.

15. The method of claim 14, wherein said second time to rebuild is determined in accordance with one or more factors including any of performance characteristics of physical drives of the first candidate storage pool configuration, workload of a data storage system directed to the first RAID group when performing said RAID-group internal rebuild processes for the first RAID group, physical location and/or availability of a spare device used in connection with performing said RAID-group internal rebuild processes for the first RAID group, a service level factor affecting an amount of time to obtain a resource in connection with performing said RAID-group internal rebuild processes, a RAID level and RAID group configuration indicated in the first candidate storage pool configuration of physical drives of the first RAID group, a priority to performing processing of incoming I/O operations over other processing of the RAID-group internal rebuild, and a rate factor related to a rate at which a particular software version is able to perform rebuild processing wherein said rate factor varies with software version.

16. The method of claim 14, wherein said first time to restore is determined in accordance with one or more factors including any of a service level factor affecting an amount of time to obtain a resource in connection with performing data restoration processing to restore data for said first candidate pool configuration, physical location of a backup data set used in connection with performing data restoration processing to restore data for said first candidate pool configuration, a type of data restoration process performed in connection with restoring data for said first candidate pool configuration, and a physical location and/or availability of a spare device.

17. A method of selecting a storage pool configuration comprising:

determining a set of one or more candidate storage pool configurations,

performing, for each of the candidate storage pool configurations of the set, a reliability calculation to determine a reliability value indicating reliability of said each candidate storage pool configuration;

,

performing, for each of the candidate storage pool configurations of the set, an availability calculation to determine an availability value indicating availability of said each candidate storage pool configuration; and

selecting, in accordance with selection criteria, one of the candidate storage pool configurations of the set for a first storage pool, said selection criteria including said availability value and said reliability value for each of said plurality of candidate storage pool configurations, and wherein said reliability calculation and said availability calculation are based on probabilities determined in accordance with a time-dependent model using time-dependent Markov solutions and said first storage pool is a pool used to provision storage for virtually provisioned devices wherein each of said virtually provisioned devices has a capacity and an amount of storage allocated for each virtually provisioned devices at a point in time is less than the capacity.

18. A method of selecting a storage pool configuration comprising:

determining a set of one or more candidate storage pool configurations;

performing, for each of the candidate storage pool configurations of the set, a reliability calculation to determine a reliability value indicating reliability of said each candidate storage pool configuration;

performing, for each of the candidate storage pool configurations of the set, an availability calculation to determine an availability value indicating availability of said each candidate storage pool configuration; and

selecting, in accordance with selection criteria, one of the candidate storage pool configurations of the set for a first storage pool, said selection criteria including said availability value and said reliability value for each of said plurality of candidate storage pool configurations, and wherein performing a reliability calculation for a first of the candidate storage pool configurations of the set includes determining a time to rebuild data for a first RAID group in said first candidate storage pool configuration upon an occurrence of an internal RAID group failure of the first RAID group where the first RAID group is able to recover desired data using RAID-group internal rebuild processes for the first RAID group and wherein said time to rebuild is determined in accordance with one or more factors including any of performance characteristics of physical drives of the first candidate storage pool configuration, workload of a data storage system directed to the first RAID group when performing said RAID-group internal rebuild processes for the first RAID group, physical location and/or availability of a spare device used in connection with performing said RAID-group internal rebuild processes for the first RAID group, a service level factor affecting an amount of time to obtain a resource in connection with performing said RAID-group internal rebuild processes, and a RAID level and RAID group configuration indicated in the first candidate storage pool configuration of physical drives of the first RAID group.

19. A non-transitory computer readable medium comprising code stored thereon for selecting a configuration for a storage pool, the non-transitory computer readable medium comprising code that, when executed by a processor, performs a method comprising:

determining a set of one or more candidate storage pool configurations;

performing, for each of the candidate storage pool configurations of the set, a reliability calculation to determine a reliability value indicating reliability of said each candidate storage pool configuration;

performing, for each of the candidate storage pool configurations of the set, an availability calculation to determine an availability value indicating availability of said each candidate storage pool configuration; and

selecting, in accordance with selection criteria, one of the candidate storage pool configurations of the set for a first storage pool, said selection criteria including said availability value and said reliability value for each of said plurality of candidate storage pool configuration, and wherein said selecting further comprises;

determining that said reliability value for said one candidate storage pool configuration represents a level of storage pool reliability that is greater than an acceptable reliability threshold; and

determining that said availability value for said one candidate storage pool configuration represents a level of storage pool availability that is greater than an acceptable availability threshold.