Method of monitoring resource usage in computing environment
원문보기
IPC분류정보
국가/구분
United States(US) Patent
등록
국제특허분류(IPC7판)
G06F-015/173
G06F-009/455
출원번호
US-0270137
(2005-11-08)
등록번호
US-8230059
(2012-07-24)
발명자
/ 주소
Santos, Jose Renato
Menon, Aravind
Turner, Yoshio Frank
Janakiraman, Gopalakrishnan
출원인 / 주소
Hewlett-Packard Development Company, L.P.
인용정보
피인용 횟수 :
35인용 특허 :
5
초록▼
An embodiment of monitoring resource usage in a computing environment begins with a first client registering a session with a monitoring entity. The session comprises an event type, a count, a start condition, an end condition, and a sample type. Next, a second client registers with the session. Upo
An embodiment of monitoring resource usage in a computing environment begins with a first client registering a session with a monitoring entity. The session comprises an event type, a count, a start condition, an end condition, and a sample type. Next, a second client registers with the session. Upon occurrence of the start condition and until occurrence of the end condition or until interrupted, the monitoring entity iteratively initiates a counting of occurrences of the event type that continues until reaching the count and saves a sample according to the sample type upon reaching the count. The resource monitoring entity provides at least a portion of a plurality of samples to the first client.
대표청구항▼
1. A method of monitoring resource usage in a computing environment comprising the steps of: a first client computing entity registering a resource monitoring session with a resource monitoring entity in the computing environment that comprises hardware, the resource monitoring session comprising an
1. A method of monitoring resource usage in a computing environment comprising the steps of: a first client computing entity registering a resource monitoring session with a resource monitoring entity in the computing environment that comprises hardware, the resource monitoring session comprising an event type, a count, a start condition, an end condition, and a sample type;a second client computing entity registering with the resource monitoring session;upon occurrence of the start condition and until occurrence of the end condition or until interrupted, the resource monitoring entity iteratively:initiating a counting of occurrences of the event type by a counter that continues until the counter reaches the count; andsaving a sample according to the sample type upon the counter reaching the count; andthe resource monitoring entity providing at least a portion of a plurality of samples to the first client computing entity, and at least a portion of the plurality of samples to the second client computing entity. 2. The method of claim 1 wherein the first and second client computing entities are selected from processes, virtual machines, nodes, Web clients, storage device clients, network interface card clients, file server clients, and database server clients. 3. The method of claim 1 wherein the event type is selected from a hardware event, a network transfer event, a Web server event, a storage device usage event, a storage device performance event, a storage device error event, a network card usage event, a network card performance event, a network card error event, a file server performance event, a file server usage event, a file server error event, a file server access violation, a database system usage event, and a database performance event. 4. The method of claim 1 wherein the resource monitoring entity is selected from an operating system, a virtual machine monitor, a switch software, a monitoring node software, Web server software, storage device software, network interface card software, file server software, and database management software. 5. The method of claim 1 wherein the sample type is selected from a code-in-execution indicator, a packet-based identifier, a protocol identifier, a packet size, a client identifier, an application identifier, a connection identifier, a meta-data identifier, a file system identifier, a lock queue length, and a database query. 6. The method of claim 1 wherein the resource monitoring entity has a higher privilege level than a privilege level for the first and second client computing entities. 7. The method of claim 1 wherein each sample includes a resource measurement, wherein the resource measurement measures one of (1) a resource usage of a resource in the computing environment by the first and second client computing entities independent of other resource usage of the resource by other client computing entities, and (2) a collective resource usage of the resource by a group of client computing entities that includes the first and second client computing entities. 8. The method of claim 1 wherein the event type and the count are a first event type and a first count and the resource monitoring session further comprises a second event type and a second count. 9. The method of claim 8 wherein, upon occurrence of the start condition, the resource monitoring entity iteratively counts occurrences of the second event type until reaching the second count and saves a second sample, thereby saving a plurality of second samples. 10. The method of claim 9 wherein the resource monitoring session further comprises a third event type and a third count. 11. The method of claim 10 wherein, upon occurrence of the start condition, the resource monitoring entity iteratively counts occurrences of the third event type until reaching the third count and saves a third sample, thereby saving a plurality of third samples. 12. The method of claim 11 wherein the resource monitoring entity maintains first, second, and third event tables for the first, second, and third client computing entities, respectively. 13. The method of claim 12 wherein a particular entry in each of the first, second, and third event tables includes an event identifier and the event type. 14. The method of claim 1 wherein the start condition comprises the second client computing entity registering with the resource monitoring session. 15. The method of claim 1 further comprising a third client computing entity registering with the resource monitoring session. 16. The method of claim 15 wherein the start condition comprises the second and third client computing entities registering with the resource monitoring session. 17. The method of claim 15 wherein the resource monitoring entity maintains an access control table that indicates permissions for the first, second, and third client computing entities to the plurality of samples. 18. The method of claim 17 wherein a privileged client computing entity has a special permission that allows the privileged client computing entity to issue an access control command that modifies the access control table. 19. The method of claim 1 wherein the resource monitoring session is a first resource monitoring session and further comprising a third client computing entity registering a second resource monitoring session with the resource monitoring entity. 20. The method of claim 19 wherein the resource monitoring entity maintains a session table of status information for the first and second monitoring sessions. 21. The method of claim 1 wherein the start condition is selected from registration of the resource monitoring session, another client computing entity joining the resource monitoring session, reaching a particular point-in-time, an appearance of a particular address, an appearance of a particular protocol, and access to a particular file system, database, or table within a database. 22. The method of claim 1 wherein the end condition is selected from notification by the first client computing entity, another client computing entity leaving the resource monitoring session, reaching a particular point-in-time, disappearance of a particular address, disappearance of a particular protocol, and closing of a particular file system, database, or table within a database. 23. A method of monitoring resource usage in a computing environment comprising the steps of: a first virtual machine registering a resource monitoring session with a virtual machine monitor, the resource monitoring session comprising a profiling event, a count, a start condition, an end condition, and a sample type;a second virtual machine registering with the resource monitoring session;upon occurrence of the start condition and until occurrence of the end condition or until interrupted, the virtual machine monitor iteratively:initiating a counting of occurrences of the profiling event by a hardware event counter that continues until the hardware event counter reaches the count; andsaving a sample according to the sample type upon the hardware event counter reaching the count; andthe virtual machine monitor providing at least a portion of a plurality of samples to the first virtual machine, and at least a portion of the plurality of samples to the second virtual machine. 24. The method of claim 23 wherein the profiling event is selected from a performance event, a debug event, an error event, and a security/violation event. 25. The method of claim 23 wherein the profiling event is selected from an executed instruction, a clock cycle, a cache event, a cache hit, a cache miss, a virtual memory translation look-aside buffer event, a virtual memory translation look-aside buffer hit, and a virtual memory translation look-aside buffer miss. 26. The method of claim 23 wherein the sample type is a code-in-execution indicator. 27. The method of claim 26 wherein the code-in-execution indicator is selected from an operating mode and a code-in-execution identifier. 28. The method of claim 27 wherein the operating mode is selected from a user mode, a kernel mode, and a hypervisor mode. 29. The method of claim 27 wherein the code-in execution identifier is selected from an instruction identifier, a group-of-instructions identifier, a function identifier, and a software module identifier. 30. A non-transitory computer readable medium comprising computer code that when executed by a computer implements a method of monitoring resource usage in a computing environment by a resource monitoring entity, the method of monitoring the resource usage by the resource monitoring entity comprising the steps of: receiving a first registration from a first client computing entity that registers a resource monitoring session with the resource monitoring entity, the resource monitoring session comprising an event type, a count, a start condition, an end condition, and a sample type;receiving a second registration from a second client computing entity that registers the second client computing entity with the resource monitoring session;upon occurrence of the start condition and until occurrence of the end condition or until interrupted, iteratively:initiating a counting of occurrences of the event type by a counter that continues until the counter reaches the count; andsaving a sample according to the sample type upon the counter reaching the count; andproviding at least a portion of a plurality of samples to the first client computing entity, and at least a portion of the plurality of samples to the second client computing entity. 31. A non-transitory computer readable medium comprising computer code that when executed by a computer implements a method of monitoring resource usage in a computing environment by a virtual machine monitor, the method of monitoring the resource usage by the virtual machine monitor comprising the steps of: receiving a first registration from a first virtual machine that registers a resource monitoring session with the virtual machine monitor, the resource monitoring session comprising a profiling event, a count, a start condition, an end condition, and a sample type;receiving a second registration from a second virtual machine that registers the second virtual machine with the resource monitoring session;upon occurrence of the start condition and until occurrence of the end condition or until interrupted, iteratively:initiating a counting of occurrences of the profiling event by a hardware event counter that continues until the hardware event counter reaches the count; andsaving a sample according to the sample type upon the hardware event counter reaching the count; andproviding at least a portion of a plurality of samples to the first virtual machine, and at least a portion of the plurality of samples to the second virtual machine. 32. The method of claim 1, wherein the resource monitoring session is to monitor usage of the hardware in the computing environment that is shared by the first and second client computing entities. 33. The method of claim 1, wherein the first client computing entity is a first virtual machine, and the second client computing entity is a second virtual machine, and wherein the resource monitoring entity is a virtual machine monitor, and wherein the counter is a hardware event counter that generates an interrupt upon reaching the count. 34. The method of claim 23, wherein the resource monitoring session is to monitor usage of computer hardware in the computing environment that is shared by the first and second virtual machines. 35. The method of claim 1, wherein the plurality of samples are saved based on iterative performance of the initiating and saving steps. 36. The method of claim 23, wherein the plurality of samples are saved based on iterative performance of the initiating and saving steps. 37. The computer readable medium of claim 30, wherein the plurality of samples are saved based on iterative performance of the initiating and saving steps. 38. The computer readable medium of claim 31, wherein the plurality of samples are saved based on iterative performance of the initiating and saving steps.
연구과제 타임라인
LOADING...
LOADING...
LOADING...
LOADING...
LOADING...
이 특허에 인용된 특허 (5)
Berry, Robert Francis; Hussain, Riaz Y.; Levine, Frank Eliot; Urquhart, Robert J., Method and system for SMP profiling using synchronized or nonsynchronized metric variables with support across multiple systems.
Cascaval, Gheorghe C.; Duesterwald, Evelyn; Sweeney, Peter F.; Wisniewski, Robert W., Method for vertical integrated performance and environment monitoring.
Horsnell, Matthew James; Emmons, Christopher Daniel, Data processing apparatus and method and method for generating performance monitoring interrupt signal based on first event counter and second event counter.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.