Method and device for identifying bottleneck candidate

• US 10,713,142 B2
• Filed: 07/12/2017
• Issued: 07/14/2020
• Est. Priority Date: 07/21/2016
• Status: Active Grant

• Current Assignee: Fujitsu Limited
• Original Assignee: Fujitsu Limited
• Inventors: Kitajima, Shinya, Kikuchi, Shinji
• Primary Examiner(s): Lau, Tung S
• Application Number:
  US15/647,295
• Publication Number:
  US 20180024905A1
• Time in Patent Office: 1,098 Days
• Field of Search: None
• CPC Class Codes: G06F 11/34 Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
• CPC Class Codes: G06F 11/3414 Workload generation, e.g. scripts, playback
• CPC Class Codes: G06F 11/3423 where the assessed time is active or idle time
• CPC Class Codes: G06F 11/3447 Performance evaluation by modeling

Abstract

A method includes, based on communication times regarding an execution command and a response of each of a plurality of services included in a task executed two or more times, and on information on an information processing device that executes each of the plurality of services, generating a group of one or more services executed continuously by a same information processing device in each of the executions of the task, calculating a first processing time of an entirety of the one or more services in each group, calculating a second processing time per service obtained by dividing the first processing time by a number of the one or more services in each group, calculating an average processing time by averaging the second processing times for each of the plurality of services for each task, and outputting a specific service based on the average processing time.

Claims

1. A method, comprising:

transmitting, by one or more information processing devices, communication time information related to processes of a plurality of services included in each of a plurality of tasks and information on the one or more information processing devices that executed the plurality of services, the communication time information including a transmission time of an execution command for each of the plurality of services and a response time corresponding to the transmission time;

receiving, by a computer, the communication time information and the information on the one or more information processing devices;

calculating, for each of the plurality of tasks, a processing time for each of a plurality of groups obtained by grouping services continuously executed in a same information processing devices of the one or more information processing devices based on the received communication time information and the received information on the one or more information processing devices;

calculating, for each of the plurality of tasks, a service processing time for each of the plurality of groups by dividing the calculated processing time for a group by a number of services included in the group, wherein the service processing time is calculated more accurately by subtracting a processing time of a service that waits for a response from a relevant processing time and wherein the service processing time is calculated more quickly by subtracting a transmission time from a reception time;

calculating, for each of the plurality of tasks, an average processing time by averaging the processing times and the service processing times;

extracting a service corresponding to an average processing time among the calculated average processing times from the plurality of services that is greater than or equal to a predetermined threshold; and

outputting, at a display, the extracted service as a bottleneck candidate when the average processing time is greater than or equal to the predetermined threshold.

2. The method according to claim 1, further comprising:

after the calculating of a first processing time, determining whether or not a service that waits for a response from another service is present in each of the groups based on response information that indicates a state where each of the plurality of services waits for a response from another service; and

when it is determined that the service that waits for a response is present, subtracting a processing time of a group including a service that transmits the response from a processing time of a group including the service that waits for the response.

3. The method according to claim 1, further comprising:

determining whether or not a service that waits for a response from another service is present based on response information that indicates a state where each of the plurality of services waits for a response;

when it is determined that the service that waits for the response is not present, receiving, from a relay device coupled to each of the one or more information processing devices, a reception time at the relay device in receiving a specific execution command to execute a first service in the group, and a transmission time at the relay device in transmitting another instruction command to execute a service to be executed subsequent to the group; and

calculating a first processing time per group by subtracting the reception time from the transmission time.

4. The method according to claim 1, further comprising:

determining whether or not a service that waits for a response from another service is present based on response information that indicates a state where each of the services waits for a response;

when it is determined that the service that waits for the response is present, receiving, from a relay device coupled to each of the one or more information processing devices, a reception time at the relay device in receiving a specific execution command to execute the services in the group including the service that waits for the response, and a transmission time at the relay device in transmitting a response command indicating a response to the execution command; and

calculating a first processing time per group by subtracting the reception time from the transmission time.

5. The method according to claim 1, further comprising:

displaying the bottleneck candidate and the average processing time of the bottleneck candidate based on the average processing times.

6. The method according to claim 1, further comprising:

outputting, as the bottleneck candidate, a service with the longest average processing time among the average processing times for the task.

7. A method executed by a computer, the method comprising:

based on communication times related to processing of a plurality of services included in each of a plurality of tasks and on information on an information processing device that executes each of the plurality of services, calculating a processing time per group for each of the tasks, the group being a group of one or more services continuously executed by a same information processing device;

storing a service processing time for each task, the service processing time obtained by dividing the calculated processing time per group by a number of the services included in the group, wherein the service processing time is calculated more accurately by subtracting a processing time of a service that waits for a response from a relevant processing time and wherein the service processing time is calculated more quickly by subtracting a transmission time from a reception time;

calculating an average processing time by averaging the service processing times for each of the tasks;

extracting a service corresponding to an average processing time among the calculated average processing times from the plurality of services that is greater than or equal to a predetermined threshold; and

outputting, at a display, the extracted service as a bottleneck candidate when the average processing time is greater than or equal to the predetermined threshold.

8. A device comprising:

a memory; and

a processor coupled to the memory and configured to;

transmit communication time information related to processes of a plurality of services included in each of a plurality of tasks and information on one or more information processing devices that executed the plurality of services, the communication time information including a transmission time of an execution command for each of the plurality of services and a response time corresponding to the transmission time,

receive the communication time information and the information on the one or more information processing devices,

calculate, for each of the plurality of tasks, a processing time for each of a plurality of groups obtained by grouping services continuously executed in a same information processing devices of the one or more information processing devices based on the received communication time information and the received information on the one or more information processing devices,

calculate, for each of the plurality of tasks, a service processing time for each of the plurality of groups by dividing the calculated processing time for a group by a number of services included in the group, wherein the service processing time is calculated more accurately by subtracting a processing time of a service that waits for a response from a relevant processing time and wherein the service processing time is calculated more quickly by subtracting a transmission time from a reception time,

calculate, for each of the plurality of tasks, an average processing time by averaging the processing times and the service processing times,

extract a service corresponding to an average processing time among the calculated average processing times from the plurality of services that is greater than or equal to a predetermined threshold, and

output, at a display, the extracted service as a bottleneck candidate when the average processing time is greater than or equal to the predetermined threshold.

9. The device according to claim 8, wherein the processor is configured to:

after a calculation of a first processing time, determine whether or not a service that waits for a response from another service is present in each of the groups based on response information that indicates a state where each of the plurality of services waits for a response from another service, and

when it is determined that the service that waits for a response is present, subtract a processing time of a group including a service that transmits the response from a processing time of a group including the service that waits for the response.

10. The device according to claim 8, wherein the processor is configured to:

determine whether or not a service that waits for a response from another service is present based on response information that indicates a state where each of the plurality of services waits for a response,

when it is determined that the service that waits for the response is not present, receive, from a relay device coupled to each of the one or more information processing devices, a reception time at the relay device in receiving a specific execution command to execute a first service in the group, and a transmission time at the relay device in transmitting another instruction command to execute a service to be executed subsequent to the group, and

calculate a first processing time per group by subtracting the reception time from the transmission time.

11. The device according to claim 8, wherein the processor is configured to:

determine whether or not a service that waits for a response from another service is present based on response information that indicates a state where each of the services waits for a response,

when it is determined that the service that waits for the response is present, receive, from a relay device coupled to each of the one or more information processing devices, a reception time at the relay device in receiving a specific execution command to execute the services in the group including the service that waits for the response, and a transmission time at the relay device in transmitting a response command indicating a response to the execution command, and

calculate a first processing time per group by subtracting the reception time from the transmission time.

12. The device according to claim 8, wherein the processor is configured to display the bottleneck candidate and the average processing time of the bottleneck candidate based on the average processing times.

13. The device according to claim 8, wherein the processor is configured to output, as the bottleneck candidate, a service with the longest average processing time among the average processing times for the task.