Method and electronic device for storage testing

Patent metadata

Field	Value
Publication number	US-12099422-B2
Application number	US-202318104553-A
Country	US
Kind code	B2
Filing date	Feb 1, 2023
Priority date	Jul 22, 2022
Publication date	Sep 24, 2024
Grant date	Sep 24, 2024

Abstract

Techniques for storage testing involve: acquiring a first state of a storage system including first input/output (IO) load information; taking a first action based on the first state, the first action causing the first IO load information to be changed to second IO load information; updating the first action to be a reserved action for the first state if it is obtained based on the second IO load information that the storage system reaches a preset condition; and obtaining an action combination of a plurality of IO load information changes based on a plurality of reserved actions corresponding to a plurality of states, wherein the plurality of states include the first state. Accordingly, the most effective load combination change mode for the storage system can be found automatically and more accurately, so as to find more vulnerabilities of the storage system, thereby improving the efficiency of storage system testing.

First claim

The invention claimed is:

1. A method for storage testing, comprising:
acquiring a first state of a storage system comprising first input/output (IO) load information;
taking a first action based on the first state, the first action causing the first IO load information to be changed to second IO load information;
updating the first action to be a reserved action for the first state if it is obtained based on the second IO load information that the storage system reaches a preset condition; and
obtaining an action combination of a plurality of IO load information changes based on a plurality of reserved actions corresponding to a plurality of states, wherein the plurality of states comprise the first state.

2. The method according to claim 1 , wherein the updating the first action to be a reserved action for the first state if it is obtained based on the second IO load information that the storage system reaches a preset condition comprises:
performing storage performance testing based on the second IO load information to obtain a first empirical value, wherein the first empirical value represents a degree of impact on the performance of the storage system by taking the first action in the first state; and
updating the first action to be the reserved action for the first state if the first empirical value is a largest value among a plurality of known empirical values of the first state, wherein the reserved action for the first state is an action that has a greatest impact on the performance of the storage system among a plurality of actions taken in the first state, the plurality of actions comprise the first action, and the action combination indicates a combination of the plurality of IO load information changes that have greatest impacts on the performance of the storage system in the plurality of states.

3. The method according to claim 2 , wherein parameters indicating the performance of the storage system comprise one or more of the following: central processing unit (CPU) utilization, memory utilization, IO latency, data miss compare, and data path failure.

4. The method according to claim 1 , wherein the reserved action for the first state is a known optimal action in the first state, and the taking a first action based on the first state comprises:
taking a random action with a first probability value and taking the reserved action for the first state with a second probability value if it is detected that the first state is a known state, wherein the first probability value and the second probability value are greater than or equal to 0 and less than or equal to 1, the first probability value is configured to decrease over time, and the second probability value is configured to increase over time; and
taking a random action if it is detected that the first state is an unknown state.

5. The method according to claim 1 , wherein the reserved action for the first state is a known optimal action in the first state, and the taking a first action based on the first state comprises:
taking the reserved action for the first state if it is detected that the first state is a known state; and
taking a random action if it is detected that the first state is an unknown state.

6. The method according to claim 1 , further comprising:
taking the first action based on the first state, and obtaining the second state, the second state comprising the second IO load information, and the plurality of states comprising the second state;
taking a second action based on the second state, the second action causing the second IO load information to be changed to third IO load information; and
updating the second action to be a reserved action for the second state if it is obtained based on the third IO load information that the storage system reaches the preset condition.

7. The method according to claim 1 , further comprising:
terminating the storage testing if it is detected that a preset testing target is achieved, and the achievement of the preset testing target comprising one or more of the following: reaching a predetermined threshold of central processing unit (CPU) utilization, reaching a predetermined threshold of memory utilization, reaching a predetermined threshold of IO latency, or discovering a critical defect comprising a data miss compare or a data path failure; and
terminating the storage testing if it is detected that a number of attempts of the first action reaches a preset maximum value.

8. The method according to claim 1 , wherein the first state further comprises one or more of the following parameters: hardware information, platform information, and drive information of a storage testing system, load type, load thread, and load profile of storage IO, and total number of input/output operations per second (IOPS), central processing unit (CPU) utilization, memory utilization, IO latency, and current event during running of the storage testing, the load profile comprising deduplication, block size, and read/write ratio, and the current event comprising no event, redundant array of independent disks (RAID) rebuilding, system space exhaustion and recovery, snapshot creation/deletion, data simplification tag on/off, synchronously triggered replication, logical unit access (LUN) switching, and storage processor restart.

9. The method according to claim 1 , wherein the taking a first action comprises one or more of the following: no action, modifying a load type, modifying a load thread, modifying a load profile, and injecting a special event, the special event comprising redundant array of independent disks (RAID) rebuilding, system space exhaustion and recovery, snapshot creation/deletion, data simplification tag on/off, synchronously triggered replication, logical unit access (LUN) switching, and storage processor restart.

10. The method according to claim 1 , wherein the storage system includes a set of storage processors constructed and arranged to perform IO operations on behalf of a set of users.

11. An electronic device for storage testing, comprising:
a processor; and
a memory coupled to the processor and having instructions stored therein, wherein the instructions, when executed by the processor, cause the electronic device to perform operations comprising:
acquiring a first state of a storage system comprising first input/output (IO) load information;
taking a first action based on the first state, the first action causing the first IO load information to be changed to second IO load information;
updating the first action to be a reserved action for the first state if it is obtained based on the second IO load information that the storage system reaches a preset condition; and
obtaining an action combination of a plurality of IO load information changes based on a plurality of reserved actions corresponding to a plurality of states, wherein the plurality of states comprise the first state.

12. The electronic device according to claim 11 , wherein the updating the first action to be a reserved action for the first state if it is obtained based on the second IO load information that the storage system reaches a preset condition comprises:
performing storage performance testing based on the second IO load information to obtain a first empirical value, wherein the first empirical value represents a degree of impact on the performance of the storage system by taking the first action in the first state; and
updating the first action to be the reserved action for the first state if the first empirical value is a largest value among a plurality of known empirical values of the first state, wherein the reserved action for the first state is an action that h

Assignees

• Dell Products Lp

Inventors

• Chen Chi
• Dai Changyue
• Shi En
• Dong Hailan

Classifications

• G06F11/2221 Primary

  to test input/output devices or peripheral units · CPC title

• G06F11/2273

  Test methods · CPC title

• G11C29/56008

  Error analysis, representation of errors · CPC title

• G06F11/221 Primary

  to test buses, lines or interfaces, e.g. stuck-at or open line faults · CPC title

• G11C29/08 Primary

  Functional testing, e.g. testing during refresh, power-on self testing [POST] or distributed testing · CPC title