Management of shared storage I/O resources

Patent metadata

Field	Value
Publication number	US-9436389-B2
Application number	US-201414529436-A
Country	US
Kind code	B2
Filing date	Oct 31, 2014
Priority date	Mar 8, 2007
Publication date	Sep 6, 2016
Grant date	Sep 6, 2016

Abstract

Automated management of shared I/O resources involves use of a policy engine for implementing I/O scheduling group I/O policies. The I/O policies are used for determining whether corresponding I/O requests should be issued to a shared storage system immediately or should be delayed via corresponding policy-based queues. In the context of database systems, a database administrator can specify policies regarding how I/O resources should be used and the database system itself enforces the policies, rather than requiring the database administrator enforce the I/O usage of the database and of the individual users.

First claim

What is claimed is:

1. A computer-executed method for managing shared storage input-output (I/O) resources, comprising:
receiving a request for an I/O operation on a storage system to which a plurality of requests for I/O operations have been recently issued;
wherein the plurality of requests comprises a first set of requests that are associated with a first I/O scheduling group of a plurality of I/O scheduling groups, and a second set of requests that are associated with a second I/O scheduling group of the plurality of I/O scheduling groups;
determining an I/O scheduling group of the plurality of I/O scheduling groups with which the received request is associated;
in response to determining that the received request is associated with the first I/O scheduling group:
identifying a first I/O policy, associated with the first I/O scheduling group, that specifies a first I/O allocation that dictates a first maximum threshold, wherein the first maximum threshold is one of:
a first maximum I/O rate at which I/O operations should issue to said storage system for requests associated with said first I/O scheduling group, or
a first maximum throughput rate at which I/O operations should issue to said storage system for requests associated with said first I/O scheduling group,

determining an I/O statistic associated with I/O operations that have recently issued to said storage system for said first I/O scheduling group,
wherein the I/O statistic is one of: I/O rate and I/O throughput;
determining, based at least in part on whether said I/O statistic indicates that said first maximum threshold has been exceeded, whether to place said request in an I/O request queue, and
in response to determining to place said request in an I/O request queue:
placing said request in a particular I/O request queue associated with said first I/O scheduling group, and
issuing said request to said storage system when said request is dequeued from said particular I/O request queue based on said first I/O policy, for performance of said I/O operation by said storage system;

wherein a second I/O policy associated with said second I/O scheduling group specifies a second I/O allocation that dictates a second maximum threshold, wherein the second maximum threshold is one of:
a second maximum I/O rate at which I/O operations should issue to said storage system for requests associated with said second I/O scheduling group, or
a second maximum throughput rate at which I/O operations should issue to said storage system for requests associated with said second I/O scheduling group;

determining whether said I/O statistic associated with I/O operations that have recently issued to said storage system for said first I/O scheduling group is less than said first maximum threshold; and
in response to determining that said I/O statistic is less than said first maximum threshold:
issuing more requests associated with said second I/O scheduling group than prescribed by said second maximum threshold;

wherein the method is performed by one or more computing devices.

2. One or more non-transitory computer-readable media storing instructions, wherein the instructions include:
instructions which, when executed by one or more hardware processors, cause receiving a request for an I/O (input/output) operation on a storage system to which a plurality of requests for I/O operations have been recently issued;
wherein the plurality of requests comprises a first set of requests that are associated with a first I/O scheduling group of a plurality of I/O scheduling groups, and a second set of requests that are associated with a second I/O scheduling group of the plurality of I/O scheduling groups;
instructions which, when executed by one or more hardware processors, cause determining an I/O scheduling group of the plurality of I/O scheduling groups with which the received request is associated;
instructions which, when executed by one or more hardware processors, cause, in response to determining that the received request is associated with the first I/O scheduling group:
identifying a first I/O policy, associated with the first I/O scheduling group, that specifies a first I/O allocation that dictates a first maximum threshold, wherein the first maximum threshold is one of:
a first maximum I/O rate at which I/O operations should issue to said storage system for requests associated with said first I/O scheduling group, or a first maximum throughput rate at which I/O operations should issue to said storage system for requests associated with said first I/O scheduling group, determining an I/O statistic associated with I/O operations that have recently issued to said storage system for said first I/O scheduling group, wherein the I/O statistic is one of: I/O rate and I/O throughput;
determining, based at least in part on whether said I/O statistic indicates that said first maximum threshold has been exceeded, whether to place said request in an I/O request queue, and
in response to determining to place said request in an I/O request queue:
placing said request in a particular I/O request queue associated with said first I/O scheduling group, and
issuing said request to said storage system when said request is dequeued from said particular I/O request queue based on said first I/O policy, for performance of said I/O operation by said storage system;
wherein a second I/O policy associated with said second I/O scheduling group specifies a second I/O allocation that dictates a second maximum threshold, wherein the second maximum threshold is one of:
a second maximum I/O rate at which I/O operations should issue to said storage system for requests associated with said second I/O scheduling group, or a second maximum throughput rate at which I/O operations should issue to said storage system for requests associated with said second I/O scheduling group;
instructions which, when executed by one or more hardware processors, cause determining whether said I/O statistic associated with I/O operations that have recently issued to said storage system for said first I/O scheduling group is less than said first maximum threshold; and
instructions which, when executed by one or more hardware processors, cause, in response to determining that said I/O statistic is less than said first maximum threshold: issuing more requests associated with said second I/O scheduling group than prescribed by said second maximum threshold.

3. The computer-executed method of claim 1 , wherein determining whether to place said request in an I/O request queue comprises performing one of:
comparing said I/O statistic to said first maximum I/O rate, wherein said I/O statistic is an I/O rate; or
comparing said I/O statistic to said first maximum throughput rate, wherein said I/O statistic is an I/O throughput.

4. The computer-executed method of claim 1 , wherein issuing said request to said storage system when said request is dequeued from said particular I/O request queue based on said first I/O policy comprises:
determining the first maximum I/O rate is not exceeded for the first I/O scheduling group; and
in response to determining the first maximum I/O rate is not exceeded for the first I/O scheduling group, dequeuing the request from the particular I/O request queue.

5. The computer-executed method of claim 1 , further comprising if determined not to place said request in an I/O request queue, then immediately issuing said request to said storage system.

6. The computer-executed method of claim 1 , wherein said request is issued to said storage system from a database server.

7. The computer-executed method of claim 1 , wherein:
the first I/O allocation is a particular percentage of the total resources, of said storage system, avai

Assignees

• Oracle Int Corp

Inventors

• Lee Sue-Kyoung
• Susairaj Margaret
• Chatterjee Sumanta

Classifications

• G06F3/0659

  Command handling arrangements, e.g. command buffers, queues, command scheduling · CPC title

• G06F3/0683

  Plurality of storage devices · CPC title

• G06F3/067

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

• G06F3/061 Primary

  Improving I/O performance · CPC title