PER-STORAGE CLASS QUALITY OF SERVICE MANAGEMENT WITHIN A DISTRIBUTED STORAGE NETWORK

• US 20180081592A1
• Filed: 09/20/2016
• Published: 03/22/2018
• Est. Priority Date: 09/20/2016
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Hegde, Harsha, Leggette, Wesley B.
• Granted Patent: US 10,558,389 B2
• Time in Patent Office: Days
• CPC Class Codes: G06F 11/1076 Parity data used in redundant arrays of independent storages, e.g. in RAID systems
• CPC Class Codes: G06F 3/0611 in relation to response time
• CPC Class Codes: G06F 3/064 Management of blocks
• CPC Class Codes: G06F 3/0659 Command handling arrangements, e.g. command buffers, queues, command scheduling
• CPC Class Codes: G06F 3/067 Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]

Abstract

A method includes receiving a request to access a set of encoded data slices from a user computing device, where the set of encoded data slices is stored in a set of storage units of a dispersed storage network (DSN). A storage container includes the set of storage units and supports a plurality of logical storage vaults. The method further includes determining a quality of service (QoS) matrix regarding the request when QoS issues exists. The QoS matrix includes classes for user computing devices, logical storage vaults, and storage containers, and a plurality of performance properties for each class. The method further includes determining QoS levels attributable to the user computing device, a first logical storage vault, and the storage container. The method further includes determining processing priority of the request based on the QoS matrix and levels and executing the request in accordance with the processing priority.

Claims

1. A method comprises:

receiving, by a dispersed storage (DS) processing unit of a dispersed storage network (DSN), a request to access a set of encoded data slices from a user computing device of a plurality of user computing devices of the DSN, wherein a data segment of a data object is dispersed storage error encoded into the set of encoded data slices, wherein the set of encoded data slices is stored in a set of storage units of the DSN, wherein a storage container of the DSN includes the set of storage units, wherein the storage container supports a plurality of logical storage vaults, and wherein the user computing device is affiliated with a first logical storage vault of the plurality of logical storage vaults;

when a quality of service (QoS) issue exists for the DSN, determining, by the DS processing unit, a QoS matrix regarding the request, wherein the QoS matrix includes classes for user computing devices, logical storage vaults, and storage containers and, for each class, includes a plurality of performance properties;

determining, by the DS processing unit, QoS levels attributable to the user computing device, the first logical storage vault, and the storage container;

determining, by the DS processing unit, processing priority of the request based on the QoS matrix and the QoS levels; and

executing, by the DS processing unit, the request in accordance with the processing priority.

2. The method of claim 1, wherein the plurality of performance properties comprises two or more of:

number of requests per time frame;

data transfer rate;

storage quota; and

latency guarantee.

3. The method of claim 1 further comprises:

determining, by the DS processing unit, that the QoS issue exists when the DSN includes one or more of insufficient storage units and insufficient DS processing units.

4. The method of claim 1 further comprises:

determining, by the DS processing unit, a container level QoS issue as the QoS issue;

determining, by the DS processing unit, whether accessing of the first logical storage vault exceeds the QoS level attributable to the first logical storage vault; and

when the accessing of the first logical storage vault exceeds the QoS level attributable to the first logical storage vault, lowering, by DS processing unit, priority of the request in a request priority queue.

5. The method of claim 4 further comprises:

when the accessing of the first logical storage vault does not exceed the QoS level attributable to the first logical storage vault, determining, by the DS processing unit, whether accessing of another logical storage vault of the plurality of logical storage vaults exceeds the QoS level attributable to the other logical storage vault;

when the accessing of the other logical storage vault exceeds the QoS level attributable to the other logical storage vault, increasing, by DS processing unit, priority of the request in the request priority queue; and

when the accessing of the other logical storage vault does not exceed the QoS level attributable to the other logical storage vault, maintaining, by DS processing unit, priority of the request in the request priority queue.

6. The method of claim 1 further comprises:

determining, by the DS processing unit, a vault level QoS issue as the QoS issue;

determining, by the DS processing unit, whether accessing by the user computing device exceeds the QoS level attributable to the user computing device; and

when the accessing by the user computing device exceeds the QoS level attributable to the user computing device, lowering, by DS processing unit, priority of the request in a request priority queue.

7. The method of claim 6 further comprises:

when the accessing by the user computing device does not exceed the QoS level attributable to the user computing device, determining, by the DS processing unit, whether accessing of another user computing device of the plurality of user computing devices exceeds the QoS level attributable to the other user computing device;

when the accessing by the other user computing device exceeds the QoS level attributable to the other user computing device, increasing, by DS processing unit, priority of the request in the request priority queue; and

when the accessing by the other user computing device does not exceed the QoS level attributable to the other user computing device, maintaining, by DS processing unit, priority of the request in the request priority queue.

8. The method of claim 1 further comprises:

determining, by the DS processing unit, a user computing device level QoS issue as the QoS issue;

determining, by the DS processing unit, whether accessing by the user computing device exceeds the QoS level attributable to the user computing device;

when the accessing by the user computing device exceeds the QoS level attributable to the user computing device, lowering, by DS processing unit, priority of the request in a request priority queue; and

when the accessing by the user computing device does not exceed the QoS level attributable to the user computing device, maintaining, by DS processing unit, priority of the request in the request priority queue.

9. A dispersed storage (DS) processing unit of a dispersed storage network (DSN) comprises:

an interface;

memory; and

a processing module operably coupled to the memory and the interface, wherein the processing module is operable to;

receive a request to access a set of encoded data slices from a user computing device of a plurality of user computing devices of the DSN, wherein a data segment of a data object is dispersed storage error encoded into the set of encoded data slices, wherein the set of encoded data slices is stored in a set of storage units of the DSN, wherein a storage container of the DSN includes the set of storage units, wherein the storage container supports a plurality of logical storage vaults, and wherein the user computing device is affiliated with a first logical storage vault of the plurality of logical storage vaults;

when a quality of service (QoS) issue exists for the DSN, determine a QoS matrix regarding the request, wherein the QoS matrix includes classes for user computing devices, logical storage vaults, and storage containers and, for each class, includes a plurality of performance properties;

determine QoS levels attributable to the user computing device, the first logical storage vault, and the storage container;

determine processing priority of the request based on the QoS matrix and the QoS levels; and

execute the request in accordance with the processing priority.

10. The DS processing unit of claim 9, wherein the plurality of performance properties comprises two or more of:

number of requests per time frame;

data transfer rate;

storage quota; and

latency guarantee.

11. The DS processing unit of claim 9, wherein the processing module further functions to:

determine that the QoS issue exists when the DSN includes one or more of insufficient storage units and insufficient DS processing units.

12. The DS processing unit of claim 9, wherein the processing module further functions to:

determine a container level QoS issue as the QoS issue;

determine whether accessing of the first logical storage vault exceeds the QoS level attributable to the first logical storage vault; and

when the accessing of the first logical storage vault exceeds the QoS level attributable to the first logical storage vault, lower priority of the request in a request priority queue.

13. The DS processing unit of claim 12, wherein the processing module further functions to:

when the accessing of the first logical storage vault does not exceed the QoS level attributable to the first logical storage vault, determine whether accessing of another logical storage vault of the plurality of logical storage vaults exceeds the QoS level attributable to the other logical storage vault;

when the accessing of the other logical storage vault exceeds the QoS level attributable to the other logical storage vault, increase priority of the request in the request priority queue; and

when the accessing of the other logical storage vault does not exceed the QoS level attributable to the other logical storage vault, maintain priority of the request in the request priority queue.

14. The DS processing unit of claim 9, wherein the processing module further functions to:

determine a vault level QoS issue as the QoS issue;

determine whether accessing by the user computing device exceeds the QoS level attributable to the user computing device; and

when the accessing by the user computing device exceeds the QoS level attributable to the user computing device, lower priority of the request in a request priority queue.

15. The DS processing unit of claim 14, wherein the processing module further functions to:

when the accessing by the user computing device does not exceed the QoS level attributable to the user computing device, determine whether accessing of another user computing device of the plurality of user computing devices exceeds the QoS level attributable to the other user computing device;

when the accessing by the other user computing device exceeds the QoS level attributable to the other user computing device, increase priority of the request in the request priority queue; and

when the accessing by the other user computing device does not exceed the QoS level attributable to the other user computing device, maintain priority of the request in the request priority queue.

16. The DS processing unit of claim 9, wherein the processing module further functions to:

determine user computing device level QoS issue as the QoS issue;

determine whether accessing by the user computing device exceeds the QoS level attributable to the user computing device;

when the accessing by the user computing device exceeds the QoS level attributable to the user computing device, lower priority of the request in a request priority queue; and

when the accessing by the user computing device does not exceed the QoS level attributable to the user computing device, maintain priority of the request in the request priority queue.