Method and apparatus for using system traces to characterize workloads in a data storage system

• US 6,269,410 B1
• Filed: 02/12/1999
• Issued: 07/31/2001
• Est. Priority Date: 02/12/1999
• Status: Expired due to Term

• Current Assignee: Hewlett Packard Enterprise Development LP (Hewlett-Packard Enterprise Company)
• Original Assignee: Hewlett-Packard Company (HP Inc.)
• Inventors: Spasojevic, Mirjana
• Primary Examiner(s): Shin, Christopher B.
• Application Number:
  US09/249,399
• Time in Patent Office: 900 Days
• Field of Search: 709/102-107, 710/15-19, 711/114, 714/47
• US Class Current: 710/5
• CPC Class Codes: G06F 11/3485 for I/O devices
• CPC Class Codes: G06F 13/385 for adaptation of a particular data processing system to different peripheral devices

Abstract

I/O activity of a data storage system is characterized by collecting system traces generated during the I/O activity, grouping records in the system traces according to stores, identifying I/O activity in streams corresponding to the stores, and processing the groups of records to characterize I/O activity for the streams. The stores represent units of storage such as logical drives, single data storage devices and groups of data storage devices in the data storage system. Characterizations of the I/O activity may be used to re-allocate data across the data storage system.

Claims

1. A method of characterizing I/O activity in a data storage system, a computer generating system traces during I/O operations with the data storage system, the method comprising:

having the computer to gather the system traces;

grouping records in the gathered system traces, the records being grouped according to stores, the stores representing units of storage in the data storage system;

identifying I/O activity in streams corresponding to the stores; and

processing the groups of records to characterize I/O activity for the streams.

2. The method of claim 1, further comprising re-allocating data across the data storage system, the re-allocation being based on the I/O activity characterization.

3. The method of claim 1, wherein the traces that have been gathered are stored on a separate data storage device.

4. The method of claim 1, wherein the processing includes generating attributes for the streams, the attributes being statistics of I/O activity distribution.

5. The method of claim 1, wherein the I/O activity of a given stream is identified by examining timestamps in the trace records corresponding to the given stream, the time stamps indicating on and off phases of the given stream.

6. The method of claim 5, wherein the beginning of a phase in the stream is marked by a time stamp following a given period of inactivity, and wherein the end of the phase is marked by the last time stamp prior to a next period of I/O inactivity.

7. The method of claim 1, wherein the processing includes generating phasing attributes for each stream.

8. The method of claim 7, wherein the phasing attributes include a first attribute indicating simultaneous activity between first and second streams, the first attribute representing a length of time that the first and second streams are on together immediately after the first stream comes on, as a fraction of the length of time the first stream stays on.

9. The method of claim 1, wherein the processing includes adding record information to a running sum and examining time stamps as each record is processed, beginnings and endings of ON phases of a stream being identified by the time stamps, whereby statistics can be generated from the running counts and records can be processed only once.

10. A method of determining an overlap of I/O activity between first and second streams in a data storage system, the method comprising:

process system traces to identify I/O activity in each stream; and

determining a length of time that the first and second streams are on together immediately after the first stream comes on, as a fraction of the length of time the first stream stays on, whereby the fraction provides a measure of the overlap.

11. A system comprising

a data storage system;

first means for conducting I/O operations with the data storage devices, system traces being generated during the I/O operations;

second means for gathering the system traces;

third means for grouping records in the gathered system traces, the records being grouped according to stores, the stores being units of within the data storage system;

fourth means for identifying I/O activity in streams corresponding to the stores; and

fifth means for processing the groups of records to characterize I/O activity for the streams.

12. The system of claim 11, further comprising a separate data storage device for storing the traces that are gathered.

13. The system of claim 11, further comprising means for re-allocating data across the data storage system, the re-allocation being based on the I/O activity characterization.

14. A computer having access to system traces from a data storage system, the computer comprising at least one processor;

and computer memory;

the memory being encoded with a plurality of executable instruction, the instructions, when executed, instructing the computer to;

group records in the system traces, the records being grouped according to stores, the stores being units of storage in the data storage system;

identify I/O activity in streams corresponding to the stores; and

process the groups of records to characterize I/O activity for the streams.

15. The computer of claim 14, wherein the I/O activity of a given stream is identified by examining timestamps in the records of a store corresponding to the given stream, the time stamps indicating on and off phases of the given stream.

16. The computer of claim 15, wherein the beginning of a phase in the given stream is marked by a time stamp following a given period of inactivity, and wherein the end of the phase is marked by the last time stamp prior to a next period of I/O inactivity.

17. The computer of claim 14, wherein the processing includes generating phasing attributes for each stream.

18. The computer of claim 17, wherein the phasing attributes include a first attribute indicating simultaneous activity between first and second streams, the first attribute representing a length of time that the first and second streams are on together immediately after the first stream comes on, as a fraction of the length of time the first stream stays on.

19. The computer of claim 14, wherein the processing includes adding record information to a running sum and examining time stamps as each record is processed, beginnings and endings of ON phases of a stream being identified by the time stamps, whereby statistics can be generated from the running counts and records can be processed only once.

20. An article of manufacture for a computer having access to a multitude of system traces from a data storage system, the article comprising:

computer memory; and

a plurality of executable instructions encoded in the memory, the instructions, when executed, instructing the computer to group records in the system traces, the records being grouped according to stores, the stores being units of storage in the data storage system;

identify I/O activity in streams corresponding to the stores; and

process the groups of records to characterize I/O activity for the streams.

21. The article of claim 20, wherein the I/O activity of a given stream is identified by examining timestamps in the records of a store corresponding to the given stream, the time stamps indicating on and off phases of the given stream.

22. The article of claim 21, wherein the beginning of a phase in the given stream is marked by a time stamp following a given period of inactivity, and wherein the end of the phase is marked by the last time stamp prior to a next period of I/O inactivity.

23. The article of claim 20, wherein the processing includes generating phasing attributes for each stream.

24. The article of claim 23, wherein the phasing attributes include a first attribute indicating simultaneous activity between first and second streams, the first attribute representing a length of time that the first and second streams are on together immediately after the first stream comes on, as a fraction of the length of time the first stream stays on.

25. The article of claim 20, wherein the processing includes adding record information to a running sum and examining time stamps as each record is processed, beginnings and endings of ON phases of a stream being identified by the time stamps, whereby statistics can be generated from the running counts and records can be processed only once.