IPC분류정보
국가/구분 |
United States(US) Patent
등록
|
국제특허분류(IPC7판) |
|
출원번호 |
US-0049254
(2016-02-22)
|
등록번호 |
US-9619273
(2017-04-11)
|
발명자
/ 주소 |
- Lao, Zai Ming
- Miao, Liang
- Yu, Hai Long
|
출원인 / 주소 |
- International Business Machines Corporation
|
대리인 / 주소 |
|
인용정보 |
피인용 횟수 :
0 인용 특허 :
14 |
초록
▼
A task comprising at least a first subtask on a computer system is analyzed. As part of the analysis, a profiler controller transmits a first profile instruction to a first profiler instance. The profiler controller transmits a second profile instruction to a second profiler instance. In response to
A task comprising at least a first subtask on a computer system is analyzed. As part of the analysis, a profiler controller transmits a first profile instruction to a first profiler instance. The profiler controller transmits a second profile instruction to a second profiler instance. In response to the first profile instruction, the first profiler instances embeds a first task identifier into a first subtask request sent by a first virtual machine. In response to the second profile instruction, the second profiler instance captures the first task identifier from the first subtask request received by a second process virtual machine. The profiler controller identifies an execution instance of the task of the computer system based on the first copy and the second copy of the first task identifier received from the first profiler instance and the second profiler instance, respectively.
대표청구항
▼
1. A method of analyzing a task on a computer system, the task comprising at least a first subtask, the method comprising: transmitting, by a profiler controller to a first profiler instance, a first profile instruction to profile a first process virtual machine;transmitting, by the profiler control
1. A method of analyzing a task on a computer system, the task comprising at least a first subtask, the method comprising: transmitting, by a profiler controller to a first profiler instance, a first profile instruction to profile a first process virtual machine;transmitting, by the profiler controller to a second profiler instance, a second profile instruction to profile a second process virtual machine;embedding, by the first profiler instance and in response to the first profile instruction, a first task identifier into a first subtask request, the first subtask request sent by the first process virtual machine;receiving, by the profiler controller, a first copy of the first task identifier from the first profiler instance;capturing, by the second profiler instance and in response to the second profile instruction, the first task identifier from the first subtask request, the first subtask request received by the second process virtual machine;receiving, by the profiler controller, a second copy of the first task identifier from the second profiler instance; andidentifying, by the profiler controller, an execution instance of the task of the computer system based on the first copy of the first task identifier and the second copy of the first task identifier. 2. The method of claim 1, wherein the identifying the execution instance includes generating a report related to performance of the task of the computer system. 3. The method of claim 2, wherein the report includes a second subtask that called the first subtask, the second subtask executed by the first process virtual machine, and wherein the report further includes a third subtask that was called by the first subtask. 4. The method of claim 3, wherein the third subtask was executed by the second process virtual machine. 5. The method of claim 1 further comprising: collecting, by the first profiler instance, a first set of profiler data related to the first process virtual machine;receiving, by the profiler controller, the first set of profiler data from the first profiler instance;collecting, by the second profiler instance, a second set of profiler data related to the second process virtual machine;receiving, by the profiler controller, the second set of profiler data from the second profiler instance;determining, by the profiler controller, a performance characteristic based on the first set of profiler data and the second set of profiler data. 6. The method of claim 5, wherein the performance characteristic is selected from the group consisting of time for processing the task, time for processing subtasks related to the task, patterns of subtask calls that affect performance of the task, patterns of subtask execution that affect performance of the task, datasets that affect performance of the task, alterations of datasets based on execution of the task, and performance of the computer system based on concurrency of multiple tasks. 7. The method of claim 1 further comprising: associating, by the second profiler instance, a second subtask with the first task identifier;collecting, by the second profiler instance, a first set of profiler data related to the second subtask;receiving, by the profiler controller, the first set of profiler data and a third copy of the first task identifier from the second profiler instance; andassociating, by the profiler controller, the second subtask with the execution instance of the task of the computer system based on the first set of profiler data and the third copy of the first task identifier. 8. The method of claim 7, wherein the second subtask is executed by the second process virtual machine. 9. The method of claim 1, wherein the execution instance is provided to an end user. 10. The method of claim 1, wherein the first profiler instance and the profiler controller are located on the same physical computer. 11. The method of claim 1, wherein the first process virtual machine and the second process virtual machine are located on the same physical computer.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.