Method and systems for simulating a workload of a storage system

Patent metadata

Field	Value
Publication number	US-10001929-B2
Application number	US-201414471704-A
Country	US
Kind code	B2
Filing date	Aug 28, 2014
Priority date	Aug 30, 2013
Publication date	Jun 19, 2018
Grant date	Jun 19, 2018

Abstract

A computer system that simulates a workload of a storage system including a non-transitory storage medium including instruction code for the computer system, wherein the instruction code, when executed on the computer system, performs steps as follows: obtaining configuration data regarding first components of the first storage system including at least one RAID group; obtaining performance data including workload information of the first storage system; configuring second components of a second storage system including at least one corresponding RAID group corresponding to the at least one RAID group of the first storage system, based on the obtained configuration data; and operating the second storage system using a simulated workload based on the obtained performance data, resulting in a front loading of the second storage system equivalent to a front loading of the first storage system.

First claim

We claim:

1. A computer system that simulates a workload of a storage system comprising a non-transitory storage medium comprising instruction code for the computer system, wherein the instruction code, when executed on the computer system, performs steps as follows:
obtaining, from a first storage system, configuration data regarding first components of the first storage system including at least one first redundant array of independent disks (RAID) group, wherein the first storage system is deployed in a customer environment;
obtaining, from the first storage system, performance data comprising workload information of the first storage system;
configuring second components of a second storage system including at least one corresponding second RAID group corresponding to the at least one first RAID group, based on the obtained configuration data, wherein
the second storage system is deployed in a test environment,
a number of the second components of the second storage system is smaller than a number of the first components of the first storage system, and
only first RAID groups having a workload in excess of a predefined threshold are mapped to correspond second RAID groups; and
operating the second storage system using a simulated workload based on the obtained performance data, resulting in a front loading of the second storage system equivalent to a front loading of the first storage system.

2. The computer system according to claim 1 , wherein obtaining configuration data comprises obtaining and analyzing a dump file of the first storage system, wherein data of the dump file is representative of the configuration of the first storage system.

3. The computer system according to claim 1 , wherein the instruction code, when executed on the computer system, performs:
analyzing the obtained performance data and generating fingerprint data comprising the workload information for the simulated workload.

4. The computer system according to claim 3 , wherein the fingerprint data comprises, for each storage volume of the first storage system, a first fingerprint and/or at least one second fingerprint wherein the first fingerprint comprises an identifier of the storage volume and information about the rate, size and/or type of accesses to the storage volume including at least a number of read input/output operations per second (IOPS), a read block size, a number of write TOPS or a write block size of the analyzed performance data, and each second fingerprint comprises an identifier of the storage volume and information about an average access rate to the storage volume for a predetermined time segment analyzed comprising an average sum of read and write IOPS for the predetermined time segment of analyzed performance data, respectively.

5. The computer system according to claim 1 , wherein, in obtaining configuration data regarding first components, configuration data of at least one physical hardware component of the first storage system is obtained, and, in configuring second components, a corresponding virtual hardware component of the test environment is configured.

6. A method simulating a workload of a storage system comprising:
obtaining, from a first storage system, configuration data regarding first components of the first storage system including at least one first redundant array of independent disks (RAID) group, wherein the first storage system is deployed in a customer environment;
obtaining, from the first storage system, performance data comprising workload information of the first storage system;
configuring second components of a second storage system including at least one corresponding second RAID group corresponding to the at least one first RAID group, based on the obtained configuration data, wherein
the second storage system is deployed in a test environment,
a number of the second components of the second storage system is smaller than a number of the first components of the first storage system, and
only first RAID groups having a workload in excess of a predefined threshold are mapped to corresponding second RAID groups; and
operating the second storage system using a simulated workload based on the obtained performance data.

7. The method according to claim 6 , wherein the first components comprise a first plurality of ports, the configuration data comprises information about at least one port group and, for each port group, information about a first group of ports of the first plurality of ports associated with the respective port group, and the second components of the second storage system comprise a second plurality of ports, wherein the second storage system is configured to create only those port groups comprising at least one port of the second plurality of ports and to associate only ports of the second plurality of ports with a corresponding port group of the second storage system.

8. The method according to claim 6 , wherein the first components further comprise a first set of storage devices associated with the at least one first RAID group, the configuration data comprises data indicating a RAID level used by the at least one first RAID group, and the second components further comprise a second set of storage devices associated with the at least one second RAID group, wherein the second set of storage devices consists of a minimum number of storage devices required to simulate a RAID group using the RAID level indicated by the configuration data.

9. The method according to claim 6 , wherein obtaining configuration data comprises obtaining and analyzing a dump file of the first storage system, wherein data of the dump file is representative of the configuration of the first storage system.

10. The method according to claim 6 , further comprising analyzing the obtained performance data to generate fingerprint data comprising the workload information for the simulated workload.

11. The method according to claim 10 , wherein the fingerprint data comprises, for each storage volume of the first storage system, a first fingerprint and/or at least one second fingerprint wherein the first fingerprint comprises an identifier of the storage volume and information about the rate, size and/or type of accesses to the storage volume including at least a number of read input/output operation per second (IOPS), a read block size, a number of write IOPS or a write block size of the analyzed performance data, and each second fingerprint comprises an identifier of the storage volume and information about an average access rate to the storage volume for a predetermined time segment analyzed comprising an average sum of read and write IOPS for the predetermined time segment of analyzed performance data, respectively.

12. The method according to claim 6 , wherein, in obtaining configuration data regarding first components, configuration data of at least one physical hardware component of the first storage system is obtained, and, in configuring second components, a corresponding virtual hardware components of the test environment is configured.

13. The method according claim 12 , wherein the first storage system comprises a plurality of first RAID groups and the test environment comprises a corresponding number of second RAID groups, wherein for each first RAID group is a virtual machine controlled by a common hypervisor of the test environment and is created by cloning a virtual machine from a template virtual machine.

14. The method according to claim 6 , wherein data of the simulated workload in the second storage system follows a path corresponding to a path of data in the first storage system such that the front loading

Assignees

• Fujitsu Ltd

Inventors

• Chopra Ashish
• BINDER JüRGEN
• Montag Rainer

Classifications

• G06F3/0632

  by initialisation or re-initialisation of storage systems · CPC title

• G06F2009/4557

  Distribution of virtual machine instances; Migration and load balancing · CPC title

• G06F3/0613 Primary

  in relation to throughput · CPC title

• G06F11/3414

  Workload generation, e.g. scripts, playback · CPC title

• G06F11/3409

  for performance assessment · CPC title