Throttling in storage systems

• US 7,127,568 B2
• Filed: 01/23/2003
• Issued: 10/24/2006
• Est. Priority Date: 01/23/2003
• Status: Expired due to Term

• Current Assignee: Ceradyne, Inc. (3M Company), Google LLC (Alphabet Inc.)
• Original Assignee: Hitachi, Ltd.
• Inventors: Watanabe, Naoki, Kishiro, Shigeru
• Primary Examiner(s): Bataille, Pierre
• Assistant Examiner(s): TSAI, SHENG JEN
• Application Number:
  US10/351,053
• Publication Number:
  US 20040148484A1
• Time in Patent Office: 1,370 Days
• Field of Search: 711/170, 711/154, 711/113, 709/235
• US Class Current: 711/154
• CPC Class Codes: G06F 12/0866 for peripheral storage systems, e.g. disk cache
• CPC Class Codes: G06F 3/061 Improving I/O performance
• CPC Class Codes: G06F 3/0656 Data buffering arrangements
• CPC Class Codes: G06F 3/0659 Command handling arrangements, e.g. command buffers, queues, command scheduling
• CPC Class Codes: G06F 3/0689 Disk arrays, e.g. RAID, JBOD

Abstract

A method is provided for improving the performance of copy operations in storage systems. The method includes storing a measure of relative availability of storage system resources, detecting operations when sequential portions of a storage media are to be accessed for writing of information, and when the measure of relative availability of system resources drops to a specified extent, introducing a wait into the operation in which sequential portions of a storage media are to be accessed for writing of information. In another implementation, a method is provided for controlling writing of data in a storage system in which a step is performed of analyzing a requested write operation to determine if the requested write operation calls for a sequential task or a random task. If the requested write operation is a sequential task, information about it is stored in a task management table. the table is used to determine if throttling is needed to carry out the requested write operation, and invokes throttling if it is needed.

Claims

1. A storage system method comprising:

receiving a storage command from a host computer;

determining whether execution of writing information associated with the storage command into a cache memory is to be delayed, the determining being based on a ratio between the number storage commands that are sequential accesses and the number storage commands that are random accesses;

if execution of writing information into the cache memory is determined to be delayed, then delaying execution of writing information associated with the storage command into the cache memory;

subsequent to delaying execution, writing the information associated with the storage command into the cache memory; and

subsequent to writing to the cache memory, and further in response to receiving the storage command, storing the information in storage volumes in the storage system.

2. A method as in claim 1 wherein the measure of relative availability of the cache memory is stored in a table based on one of a group of storage volumes or a sequential task.

3. A method as in claim 2 wherein the table includes measures of allocated capacity and current capacity.

4. A method as in claim 2 wherein the table includes a measure of cache memory size available and throughput available.

5. A method as in claim 1 wherein the wait has a predetermined duration.

6. A method as in claim 1 wherein the duration of the wait is determined dynamically based upon the then available system resources.

7. In a storage system having storage volumes, the system operating under computer control and coupled to a host to receive information to be stored in the storage volumes, a method comprising:

in the storage system, receiving data I/O commands and performing I/O with the storage volumes to service the data I/O commands;

in the storage system, monitoring the data I/O commands sent between the host and the storage system to determine a measure of throughput requested by the host in having the storage system write data within the storage volumes in the storage system; and

sending to the host a request to wait if the measure of throughput exceeds a specified quantity,

wherein the measure of throughput is a rate of change of growth in cache memory usage.

8. A method as in claim 7 wherein the data to be written in the storage system is first supplied to a cache memory within the storage system, and later written to the storage volumes in the storage system.

9. A method as in claim 7 further comprising also storing the information in a secondary storage system.

10. A method as in claim 9 wherein the information is stored in a secondary storage system synchronously with being written to the storage volumes.

11. A method as in claim 9 wherein the information is stored in a secondary storage system asynchronously with being written to the storage volumes.

12. A method as in claim 7 wherein the storage volumes comprises an array of hard disk drives.

13. A method for controlling writing of data in a storage system coupled to a host computer, the storage system having a plurality of storage volumes into which the data is stored, the method comprising:

receiving I/O operations from the host, and in response thereto, performing I/O with the storage volumes to service the I/O operations;

analyzing a write operation requested by the host to determine if the requested write operation calls for a sequential task in which consecutive sectors of storage volumes are addressed or a random task in which disparate sectors of storage volumes are addressed;

if the requested write operation is a sequential task, storing information about the sequential task in a task management table;

using information in the task management table to determine if throttling is needed to carry out the requested write operation, wherein the determination is based on a ratio between the number I/O operations that are sequential accesses and the number I/O operations that are random accesses, or based on a rate of change of growth in cache memory usage; and

if throttling is needed, applying a throttling technique to the sequential task whereby data associated with the sequential task is written to a cache memory of the storage system after a delay.

14. A method as in claim 13 wherein the determination of whether the requested write operation calls for a sequential task includes determination of whether the requested write operation involves a single group of storage volumes.

15. A method as in claim 13 wherein the determination of whether the requested write operation calls for a sequential task includes determination of whether the requested write operation involves more than a single group of storage volumes.

16. A method as in claim 15 wherein throttling is controlled based upon both the determination of a sequential task and whether more than a single group of storage volumes are to have data written to them.

17. A method as in claim 13 wherein the throttling technique is selected based upon data in a table of statistical data.

18. A method as in claim 13 wherein the throttling technique comprises inserting wait states into the write operation to delay it.

19. A method as in claim 18 wherein the step of inserting wait states into the write operation comprises inserting a predefined wait time into the write operation.

20. A method as in claim 18 wherein the step of inserting wait states into the write operation comprises calculating a wait time at the time the wait time is to be inserted into the write operation.

21. A method as in claim 18 wherein following use of the throttling technique normal operation is resumed to write any remaining data into the storage system.

22. A method as in claim 13 wherein the throttling technique comprises:

monitoring a cache memory into which data is stored before being written to the storage volumes; and

when a predetermined amount of the cache memory has been filled, waiting for the cache memory to have more space before additional data are stored in the cache memory.

23. A method as in claim 22 wherein following use of the throttling technique normal operation is resumed to write any remaining data into the storage system.

24. A method as in claim 13 wherein the table uses a measure of throughput to determine whether throttling is to be applied.

25. A method as in claim 14 wherein the table uses an indication of an amount of cache memory available to determine whether throttling is to be applied.

26. A method as in claim 14 wherein the table uses both a measure of throughput and an indication of an amount of cache memory available to determine whether throttling is to be applied.