Data services scheduling in heterogeneous storage environments

Patent metadata

Field	Value
Publication number	US-10509593-B2
Application number	US-201715662558-A
Country	US
Kind code	B2
Filing date	Jul 28, 2017
Priority date	Jul 28, 2017
Publication date	Dec 17, 2019
Grant date	Dec 17, 2019

Abstract

A method for scheduling data services in a heterogeneous storage environment is disclosed. In one embodiment, such a method includes instrumenting, in a heterogeneous storage environment, each of a plurality of storage systems to generate events when operations are performed thereon. The events are detected and indexed in a single namespace. These events may then be analyzed to generate a statistical model of I/O activity occurring in the heterogeneous storage environment over a period of time. From the statistical model, the method determines periods of reduced I/O workload across the heterogeneous storage environment. The method then schedules data services to occur during these periods of reduced I/O workload. A corresponding system and computer program product are also disclosed.

First claim

The invention claimed is:

1. A method for scheduling data services in a heterogeneous storage environment, the method comprising:
instrumenting, in a heterogeneous storage environment comprising a plurality of storage systems, each of the storage systems to generate events when operations are performed thereon;
detecting the events from the plurality of storage systems and indexing the events in a single namespace;
analyzing the events in the namespace to generate a statistical model that describes I/O activity occurring in the heterogeneous storage environment over a period of time;
determining, from the statistical model, periods of reduced I/O workload across the heterogeneous storage environment as a whole; and
scheduling data services to occur in the heterogeneous storage environment during the periods of reduced I/O workload.

2. The method of claim 1 , wherein the data services comprise at least one of: bulk loading of storage system metadata, migration of data between different storage tiers, migration of data between different storage systems, and performing deep inspection of data residing on the storage systems.

3. The method of claim 1 , wherein scheduling the data services further comprises taking into account execution times of the data services.

4. The method of claim 3 , wherein the execution times are based on actual observed execution times of the data services.

5. The method of claim 3 , wherein the execution times are default execution times.

6. The method of claim 1 , wherein scheduling the data services further comprises deferring execution of the data services in the event unexpected activities occur during the periods of reduced I/O workload in which the data services are scheduled.

7. The method of claim 1 , wherein the period of time is configurable by a user.

8. A computer program product for scheduling data services in a heterogeneous storage environment, the computer program product comprising a non-transitory computer-readable storage medium having computer-usable program code embodied therein, the computer-usable program code configured to perform the following when executed by at least one processor:
instrument, in a heterogeneous storage environment comprising a plurality of storage systems, each of the storage systems to generate events when operations are performed thereon;
detect the events from the plurality of storage systems and index the events in a single namespace;
analyze the events in the namespace to generate a statistical model that describes I/O activity occurring in the heterogeneous storage environment over a period of time;
determine, from the statistical model, periods of reduced I/O workload across the heterogeneous storage environment as a whole; and
schedule data services to occur in the heterogeneous storage environment during the periods of reduced I/O workload.

9. The computer program product of claim 8 , wherein the data services comprise at least one of: bulk loading of storage system metadata, migration of data between different storage tiers, migration of data between different storage systems, and performing deep inspection of data residing on the storage systems.

10. The computer program product of claim 8 , wherein scheduling the data services further comprises taking into account execution times of the data services.

11. The computer program product of claim 10 , wherein the execution times are based on actual observed execution times of the data services.

12. The computer program product of claim 10 , wherein the execution times are default execution times.

13. The computer program product of claim 8 , wherein scheduling the data services further comprises deferring execution of the data services in the event unexpected activities occur during the periods of reduced I/O workload in which the data services are scheduled.

14. The computer program product of claim 8 , wherein the period of time is configurable by a user.

15. A system for scheduling data services in a heterogeneous storage environment, the system comprising:
at least one processor;
at least one memory device operably coupled to the at least one processor and storing instructions for execution on the at least one processor, the instructions causing the at least one processor to:
instrument, in a heterogeneous storage environment comprising a plurality of storage systems, each of the storage systems to generate events when operations are performed thereon;
detect the events from the plurality of storage systems and index the events in a single namespace;
analyze the events in the namespace to generate a statistical model that describes I/O activity occurring in the heterogeneous storage environment over a period of time;
determine, from the statistical model, periods of reduced I/O workload across the heterogeneous storage environment as a whole; and
schedule data services to occur in the heterogeneous storage environment during the periods of reduced I/O workload.

16. The system of claim 15 , wherein the data services comprise at least one of: bulk loading of storage system metadata, migration of data between different storage tiers, migration of data between different storage systems, and performing deep inspection of data residing on the storage systems.

17. The system of claim 15 , wherein scheduling the data services further comprises taking into account execution times of the data services.

18. The system of claim 17 , wherein the execution times are based on actual observed execution times of the data services.

19. The system of claim 15 , wherein scheduling the data services further comprises deferring execution of the data services in the event unexpected activities occur during the periods of reduced I/O workload in which the data services are scheduled.

20. The system of claim 15 , wherein the period of time is configurable by a user.

Assignees

• IBM

Inventors

• Bhagwat Deepavali M
• Bhosale Nilesh P
• Dain Joseph W
• Hewitt James
• Sawdon Wayne A

Classifications

• G06F3/0653

  Monitoring storage devices or systems · CPC title

• G06F3/0685

  Hybrid storage combining heterogeneous device types, e.g. hierarchical storage, hybrid arrays · CPC title

• G06F3/061

  Improving I/O performance · CPC title

• G06F9/505

  considering the load · CPC title

• G06F3/0659 Primary

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