Redundancies for reconstruction in mass data storage systems

Patent metadata

Field	Value
Publication number	US-9541978-B2
Application number	US-201514720031-A
Country	US
Kind code	B2
Filing date	May 22, 2015
Priority date	Jun 13, 2014
Publication date	Jan 10, 2017
Grant date	Jan 10, 2017

Abstract

A mass data storage system includes a redundancy manager that uses a physical position map to select a subset of storage resources having a physical distribution satisfying at least one resource distribution rule. The physical position map identifies physical positions of storage resources relative to a number of power supply units. A read/write manager writes data redundancies to select storage resources of the selected subset to provide a predetermined level of data protection that allows for reconstruction of lost data in a number of diverse circumstances.

First claim

What is claimed is:

1. A system comprising:
a physical position map that defines physical positions of a plurality of storage resources relative to a number of power supply units;
an on-line/off-line monitor that detects a failed storage resource of the plurality of storage resources; and
a restoration manager executable by one or more processors that identifies a subset of the storage resources storing data redundancies based on the physical position map and restores data of the failed storage resource from at least one other storage resource of the identified subset powered by a different power supply unit than the failed storage resource.

2. The system of claim 1 , wherein the identified subset of the storage resources are distributed between different field replaceable units.

3. The system of claim 1 , wherein the restoration manager identifies the subset of storage resources by accessing metadata corresponding to a physical position of the failed resource.

4. The system of claim 1 , wherein no more than a predetermined number of storage resources of the identified subset share an independent power supply unit.

5. The system of claim 1 , wherein no more than a predetermined number of storage resources of the identified subset are included in a same field replaceable unit.

6. The system of claim 1 , wherein the restoration manager reconciles outdated data of the failed resource with updated data of the at least one storage resource of the identified subset.

7. The system of claim 1 , wherein the restoration manager is configured to restore data of multiple storage resources of a same field replaceable unit responsive to a power supply failure.

8. A system comprising:
a physical position map identifying physical positions of a plurality of storage resources relative to a number of power supply units;
a redundancy manager executed by one or more processors that uses the physical position map to define a subset of the storage resources having a physical distribution satisfying at least one resource distribution rule; and
a read/write manager executed by one or more processors that writes data redundancies for incoming data to select storage resources of the defined subset to provide a predetermined level of data protection.

9. The system of claim 8 , wherein the at least one resource distribution rule specifies a maximum number of storage resources of the subset sharing an independent power supply unit.

10. The system claim 8 , wherein the at least one resource distribution rule specifies a maximum number of storage resources of the subset included in a same field replaceable unit.

11. The system of claim 8 , wherein the redundancy manager is further configured to:
create a metadata object identifying locations of each of the select storage resources of the subset; and
save the metadata object in association with the data redundancies.

12. The system of claim 11 , wherein the metadata object references positions of the physical position map corresponding to each of the select storage resources of the subset.

13. The system of claim 8 , wherein the redundancy manager is further configured to:
determine a minimum number of the storage resources in the subset on which to store the data redundancies while providing the predetermined level of data protection; and
select the storage resources of the subset based on the determination of the minimum number.

14. A method comprising:
generating a physical position map that defines physical positions of a plurality of storage resources relative to a number of power supply units;
detecting a failed storage resource;
accessing metadata based on the physical position map to identify a subset of the storage resources storing redundant data for the failed storage resource; and
restoring data of the failed storage resource from one or more storage resources of the subset.

15. The method of claim 14 , wherein the subset of the storage resources are distributed between different field replaceable units.

16. The method of claim 14 , wherein the storage resources of the subset do not share a power supply unit with the failed storage resource.

17. The method of claim 14 , wherein no more than a predetermined number of storage resources of the subset are included in a same field replaceable unit.

18. The method of claim 14 , wherein no more than a predetermined number of storage resources of the subset share a common power supply unit.

19. The method of claim 14 , wherein the metadata references positions of the physical position map corresponding to locations of stored redundancies for data of the failed storage resource.

20. The method of claim 14 , further comprising:
detecting replacement of a field replaceable unit including multiple storage resources of the plurality of storage resources;
rebuilding the data of each of the multiple storage resources of the field replaceable unit from a different a subset of storage resources defined by the physical position map, wherein no more than a predetermined number of storage resources of each of the subsets share an independent power source.

Assignees

• Seagate Technology Llc

Inventors

• Frick Guy David

Classifications

• G06F3/0689 Primary

  Disk arrays, e.g. RAID, JBOD · CPC title

• G06F11/2094

  Redundant storage or storage space (G06F11/2056 takes precedence) · CPC title

• G06F3/0658

  Controller construction arrangements · CPC title

• G06F1/3203

  Power management, i.e. event-based initiation of a power-saving mode · CPC title

• G06F1/3287

  by switching off individual functional units in the computer system · CPC title