Selecting cloud storage

• US 10,908,831 B2
• Filed: 10/25/2018
• Issued: 02/02/2021
• Est. Priority Date: 10/25/2018
• Status: Active Grant

• Current Assignee: Hewlett Packard Enterprise Development LP (Hewlett-Packard Enterprise Company)
• Original Assignee: Hewlett Packard Enterprise Development LP (Hewlett-Packard Enterprise Company)
• Inventors: Slater, Alastair, Mayo, Richard Phillip
• Primary Examiner(s): Tsai, Henry
• Assistant Examiner(s): Borromeo, Juanito
• Application Number:
  US16/171,019
• Publication Number:
  US 20200133544A1
• Time in Patent Office: 831 Days
• CPC Class Codes: G06F 11/14 Error detection or correction of the data by redundancy in operation G06F11/16 takes precedence
• CPC Class Codes: G06F 11/1453 using de-duplication of the data
• CPC Class Codes: G06F 3/0608 Saving storage space on storage systems
• CPC Class Codes: G06F 3/0641 De-duplication techniques
• CPC Class Codes: G06F 3/067 Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]

Abstract

Examples include selection of a remote object storage system for a deduplication data item. Some examples may include determination of an expected usage frequency for the deduplication data item, selection of a remote object cloud storage system based on the expected usage frequency for the data item, and provision of the deduplication data item to the selected remote object storage system for storage as an object in the selected remote object storage system.

Claims

1. A method comprising:

receiving, at a computing device from a deduplication system, a plurality of deduplication data items;

selecting a first remote object storage system to store a first one of the deduplication data items based on the first one of the deduplication data items being metadata for a deduplicated representation of a stream of data and based on respective properties of a plurality of remote object storage systems, including the first remote object storage system and a second remote object storage system;

providing a first one of the deduplication data items to the first remote object storage system for storage as an object in the first remote object storage system;

selecting the second remote object storage system, different than the first remote object storage system, to store a second one of the deduplication data items based on the second one of the deduplication data items being a collection of data chunks for a deduplicated representation of a stream of data and based on respective properties of the plurality of remote object storage systems; and

providing a second one of the deduplication data items to second first remote object storage system for storage as an object in the second remote object storage system;

wherein the respective properties of the remote object storage systems comprise at least one of respective access throughput amounts, respective access latency amounts, respective access cost amounts, or respective storage cost amounts.

8. The method of claim 1, wherein the first and second remote object storage systems differ in at least one of:

access throughput, access latency, access cost, or storage cost.

9. The method of claim 1, further comprising:

determining that the first one of the deduplication data items is metadata for a deduplicated representation of a stream of data based on an object name of the first one of the deduplication data items.

10. The method of claim 9, further comprising:

determining that the second one of the deduplication data items is a collection of data chunks for a deduplicated representation of a stream of data based on an object name of the second one of the deduplication data items.

2. A system comprising:

at least one processing resource; and

a non-transitory machine-readable storage medium comprising instructions executable by the at least one processing resource to;

select a first remote object storage system to store a first deduplication data item based on the first deduplication data item being metadata for a deduplicated representation of a stream of data and based on respective properties of a plurality of remote object storage systems, including the first remote object storage system and a second remote object storage system;

instruct the selected first remote object storage system to store the first deduplication data item as an object;

select the second remote object storage system, different than the first remote object storage system, to store a second deduplication data item based on the second deduplication data item being a collection of data chunks for a deduplicated representation of a stream of data and based on respective properties of a plurality of remote object storage systems; and

instruct the selected second remote object storage system to store the second deduplication data item as an object;

wherein the respective properties of the remote object storage systems comprise at least one of respective access throughput amounts, respective access latency amounts, respective access cost amounts, or respective storage cost amounts.

3. The system of claim 2, wherein the non-transitory machine-readable storage medium comprises instructions executable by the at least one processing resource to receive the a first and second deduplication data items from a deduplication system.

11. The system of claim 2, wherein the first and second remote object storage systems differ in at least one of:

access throughput, access latency, access cost, or storage cost.

12. The system of claim 2, wherein the machine-readable storage medium comprises instructions executable by the at least one processing resource to:

determine that the first deduplication data item is metadata for a deduplicated representation of a stream of data based on an object name of the first deduplication data items.

13. The method of claim 12, wherein the machine-readable storage medium comprises instructions executable by the at least one processing resource to:

determine that the second deduplication data item is a collection of data chunks for a deduplicated representation of a stream of data based on an object name of the second deduplication data item.

4. A non-transitory computer-readable storage medium comprising instructions that, when executed by a processing apparatus, cause the processing apparatus to:

identify a first deduplication data item to be stored in one of a plurality of remote object storage systems as metadata for a deduplicated representation of a stream of data;

select a first one of the remote object storage systems to store the first deduplication data item based on the first deduplication data item being metadata for a deduplicated representation of stream data and based on respective properties of the plurality of remote object storage systems, including the first one and a second one of the remote object storage systems;

instruct the selected first one of the remote object storage systems to store the first deduplication data item as an object;

identify a second deduplication data item to be stored in one of the remote object storage systems as a collection of data chunks for a deduplicated representation of a stream of data;

select the second one of the remote object storage systems, different than the first one of the remote object storage systems, to store the second deduplication data item based on the second deduplication data item being a collection of data chunks for a deduplicated representation of a stream of data and based on respective properties of the plurality of remote object storage systems; and

instruct the selected second one of the remote object storage systems to store the second deduplication data item as an object;

wherein the respective properties of the remote object storage systems comprise at least one of respective access throughput amounts, respective access latency amounts, respective access cost amounts, or respective storage cost amounts.

5. The non-transitory computer-readable storage medium of claim 4, wherein the first one of the remote object storage systems and the second one of the remote object storage systems differ in at least one of:

access throughput, access latency, access cost, or storage cost.

6. The non-transitory computer-readable storage medium of claim 4, comprising instructions executable by the at least one processing resource to:

determine that the first deduplication data item is metadata for a deduplicated representation of a stream of data based on an object name of the first deduplication data item.

7. The non-transitory computer-readable storage medium of claim 6, comprising instructions executable by the at least one processing resource to:

determine that the second deduplication data item is a collection of data chunks for a deduplicated representation of a stream of data based on an object name of the second deduplication data item.