Object storage power consumption optimization

Patent metadata

Field	Value
Publication number	US-9798483-B2
Application number	US-201715397862-A
Country	US
Kind code	B2
Filing date	Jan 4, 2017
Priority date	Jan 28, 2015
Publication date	Oct 24, 2017
Grant date	Oct 24, 2017

Abstract

An approach for object storage power consumption optimization in a networked storage environment is provided. In a typical embodiment, a plurality of memory devices are organized into pods into which redundantly replicated objects are intelligently placed. At least one pod is activated in a read-write configuration, such that data written to the networked storage environment is written to this pod. At least one other pod is a deactivated pod, which has at least one designated memory device in a read-only configuration and a remainder of devices in a deactivated state. Requests for data contained in this deactivated pod are serviced by the designated memory device in the read-only configuration. In some embodiments, the activated and deactivated pods are rotated based on an event.

First claim

What is claimed is:

1. A method for object storage in a networked storage environment, the method comprising the processes of:
organizing a plurality of memory devices in a networked storage environment into a plurality of pods;
selecting a first pod in the plurality of pods as an active pod in a read-write configuration;
selecting a second pod in the plurality of pods as a deactivated pod, wherein each deactivated pod comprises a designated memory device in a read-only configuration and a set of remainder memory devices in a deactivated state; and
reclassifying the second pod as the active pod and the first pod as the deactivated pod,
wherein a number of remainder memory devices in a deactivated state is at least one and at most:
S −( S/R )−( R− 1),

where S is a number of the memory devices in the plurality of memory devices and R is a number of data replicas among all memory devices in a pod of the plurality of pods, and where S is greater than R.

2. The method of claim 1 , wherein the reclassifying is in response to at least one of: a capacity threshold, a threshold incremental capacity increase, a time interval, or a device failure.

3. The method of claim 1 , further comprising the step of redesignating the designated memory device as one of the set of remainder memory devices and also redesignating one of the remainder memory devices as the designated memory device in response to at least one of: a time interval, a byte read basis, a promotion cycle, and a device failure of the designated memory device.

4. The method of claim 1 , further comprising the steps of:
storing, in response to a command to change an object in the networked storage environment that is currently being stored on the deactivated pod, a changed object on the active pod;
storing a change indicator for the object; and
removing, in response to the deactivated pod being reclassified as the active pod, the object from the reclassified pod.

5. The method of claim 1 , wherein the networked storage environment is a cloud computing environment.

6. The method of claim 1 , wherein each memory device of a pod comprises at least one unique failure domain storage location feature from each other memory device of the pod.

7. A memory storage system for storing objects in a networked storage environment, comprising:
a plurality of memory storage devices in a networked storage environment;
a set of pods into which the plurality of memory storage devices are organized, the set of pods including an active pod and a deactivated pod; and
at least one computer device that executes a set of instructions that causes the system to:
write, in response to a request to write an object, the object on the memory storage devices in the active pod;
retrieve, in response to a read request for an object stored on a deactivated pod, the object from a designated memory storage device in the deactivated pod, wherein remainder memory storage devices in the deactivated pod are deactivated; and
reclassify the active pod as the deactivated pod and the deactivated pod as the active pod,

wherein a number of remainder memory storage devices in a deactivated state is at least one and at most:
S −( S/R )−( R− 1),

where S is a number of the memory storage devices in the plurality of memory storage devices and R is a number of data replicas among all memory storage devices in a pod of the plurality of pods, and where S is greater than R.

8. The memory storage system of claim 7 , wherein the reclassifying is in response to at least one of: a capacity threshold, a threshold incremental capacity increase, a time interval, or a device failure.

9. The memory storage system of claim 7 , the computer device further comprising instructions for causing the system to redesignate the designated memory storage device as one of the set of remainder memory storage devices and also redesignate one of the remainder memory storage devices as the designated memory storage device in response to at least one of: a time interval, a byte read basis, a promotion cycle, and a device failure of the designated memory storage device.

10. The memory storage system of claim 7 , the computer device further comprising instructions for causing the system to:
store, in response to a command to change an object in the networked storage environment that is currently being stored on the deactivated pod, a changed object on the active pod;
store a change indicator for the object; and
remove, in response to the deactivated pod being reclassified as the active pod, the object from the reclassified pod.

11. The memory storage system of claim 7 , wherein the networked storage environment is a cloud computing environment.

12. The memory storage system of claim 7 , wherein each memory storage device of a pod comprises at least one unique failure domain storage location feature from each other memory storage device of the pod.

13. A computer program product for object storage in a networked storage environment, the computer program product comprising a computer readable hardware storage device, and program instructions stored on the computer readable hardware storage device, to:
organize a plurality of memory devices in a networked storage environment into a plurality of pods;
select a first pod in the plurality of pods as an active pod in a read-write configuration;
selecting a second pod in the plurality of pods as a deactivated pod, wherein each deactivated pod comprises a designated memory device in a read-only configuration and a set of remainder memory devices in a deactivated state; and
reclassify the second pod as the active pod and the first pod as the deactivated pod,
wherein a number of remainder memory devices in a deactivated state is at least one and at most:
S −( S/R )−( R− 1),

where S is a number of the memory devices in the plurality of memory devices and R is a number of data replicas among all memory devices in a pod of the plurality of pods, and where S is greater than R.

14. The computer program product of claim 13 , wherein the reclassifying is in response to at least one of: a capacity threshold, a threshold incremental capacity increase, a time interval, or a device failure.

15. The computer program product of claim 13 , the computer readable hardware storage device further comprising instructions to redesignate the designated memory device as one of the set of remainder memory devices and also redesignate one of the remainder memory devices as the designated memory device in response to at least one of: a time interval, a byte read basis, a promotion cycle, and a device failure of the designated memory device.

16. The computer program product of claim 13 , the computer readable hardware storage device further comprising instructions to:
store, in response to a command to change an object in the networked storage environment that is currently being stored on the deactivated pod, a changed object on the active pod;
store a change indicator for the object; and
remove, in response to the deactivated pod being reclassified as the active pod, the object from the reclassified pod.

17. The computer program product of claim 13 , wherein the networked storage environment is a cloud computing environment.

18. The computer program product of claim 13 , wherein each memory device of a pod comprises at least one unique failure domain storage location feature from each other memory device of the pod.

Assignees

• IBM

Inventors

• Ford Egan F
• Fork Michael J

Classifications

• G06F3/0634

  by changing the state or mode of one or more devices · CPC title

• G06F3/0619

  in relation to data integrity, e.g. data losses, bit errors · CPC title

• G06F3/067

  Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS] · CPC title

• G06F3/065

  Replication mechanisms · CPC title

• G06F3/0625 Primary

  Power saving in storage systems · CPC title