Capacity planning by transaction type

• US 8,631,401 B2
• Filed: 07/24/2007
• Issued: 01/14/2014
• Est. Priority Date: 07/24/2007
• Status: Active Grant

• Current Assignee: CA, Inc. (d/b/a CA Technologies) (Broadcom, Inc.)
• Original Assignee: CA, Inc. (d/b/a CA Technologies) (Broadcom, Inc.)
• Inventors: Bansal, Jyoti Kumar, Seidman, David Isaiah, Addleman, Mark Jacob
• Primary Examiner(s): Puente, Emerson
• Assistant Examiner(s): Sun, Charlie
• Application Number:
  US11/782,346
• Publication Number:
  US 20090031066A1
• Time in Patent Office: 2,366 Days
• Field of Search: 718/104
• US Class Current: 718/1
• CPC Class Codes: G06F 11/3409 for performance assessment
• CPC Class Codes: G06F 11/3419 by assessing time
• CPC Class Codes: G06F 11/3476 Data logging G06F11/14, G06F11/2205 take precedence
• CPC Class Codes: G06F 2201/87 Monitoring of transactions
• CPC Class Codes: G06F 2209/5019 Workload prediction
• CPC Class Codes: G06F 2209/508 Monitor
• CPC Class Codes: G06F 9/5011 the resources being hardware resources other than CPUs, Servers and Terminals
• CPC Class Codes: G06F 9/5027 the resource being a machine, e.g. CPUs, Servers, Terminals
• CPC Class Codes: G06F 9/5077 Logical partitioning of resources; Management or configuration of virtualized resources specific details on emulation or internal functioning of virtual machines G06F9/455
• CPC Class Codes: G06F 9/5083 Techniques for rebalancing the load in a distributed system

Abstract

Capacity planning is performed based on expected transaction load and the resource utilization for each expected transaction. Resource usage is determined for one or more transactions or URLs based on transaction specific and non-transaction specific resource usage. Once the resource usage for each transaction is known, the expected resource usage may be determined for an expected quantity of each transaction. The actual resources needed to meet the expected resource usage are then determined. Resources may include hardware or software, such as a central processing unit, memory, hard disk bandwidth, network bandwidth, and other computing system components. The expected resource usage for a transaction may based on the usage directly related to the transaction and usage not directly related to the transaction but part of a process associated with the performed transactions.

Claims

1. A computer implemented method, comprising:

monitoring a plurality of types of transactions that are executed by an application using one or more resources;

determining, for each of the types of transactions, a first usage of each of the one or more resources that is directly attributable to each of the plurality of types of transactions based on the monitoring; and

associating a second usage of each of the one or more resources with one or more of the plurality of types of transactions based on the monitoring, the second usage associated with non-transaction specific usage of the one or more resources, the associating comprises attributing a portion of the non-transaction specific usage of the one or more resources to each of the one or more types of transactions.

2. The computer implemented method of claim 1, further comprising:

receiving one or more requests over a network, the plurality of types of transactions performed in response to receiving the one or more requests.

3. The computer implemented method of claim 2, wherein the request is a URL request.

4. The computer implemented method of claim 1, wherein the attributing comprises determining the second usage for a first type of transaction of the plurality of types of transactions by determining a load percentage of the first type of transaction with respect to the total load of the plurality of types of transactions that use the one or more resources.

5. The computer implemented method of claim 4, wherein the transactions are performed by a computer process.

6. The computer implemented method of claim 5, wherein the computer process is a Java Virtual Machine (JVM) process.

7. The computer implemented method of claim 1, wherein said attributing a portion of the non-transaction specific usage comprises:

determining the total usage level of the one or more resources by the plurality of types of transactions during a period of time as part of a computer process.

8. The computer implemented method of claim 7, wherein said attributing a portion of the non-transaction specific usage comprises:

determining the total resource usage during the period of time with respect to the computer process.

9. The computer implemented method of claim 1, further comprising:

inserting monitoring code into the application, the monitoring code for monitoring the application at runtime and reporting performance data based on monitoring the plurality of types of transactions, wherein the first usage and the second usage is derived from the performance data.

10. The computer implemented method of claim 1, further comprising:

generating performance data in response to monitoring the plurality of types of transactions, the first usage and the second usage derived from the performance data.

11. The computer implemented method of claim 1, wherein said attributing comprises:

attributing a portion of the non-transaction specific usage of a first of the resources to a first of the types of transactions based on the percentage of direct usage of the first resource by the first type of transaction.

12. The computer implemented method of claim 1, wherein the attributing comprises assigning a second usage to a first of the types of transactions based on a load percentage of the first type of transaction with respect to the total load of the plurality of types of transactions that use the resource.

13. The computer implemented method of claim 1, wherein the plurality of types of transactions are performed during a Java Virtual Machine (JVM) process.

14. One or more non-transitory processor readable storage devices having processor readable code embodied on said processor readable storage devices, said processor readable code for programming one or more processors to:

monitor a plurality of transactions that are performed by a corresponding plurality of threads in response to requests to perform the transactions;

determine, based on the monitoring, the utilization of each of one or more computer components by the threads while performing the plurality of transactions;

attribute the utilization of each of the one or more computer components by each of the threads to the corresponding type of transaction of the plurality of transactions;

determine utilization of the one or more computer components by the types of transactions that is not specific to the threads; and

attribute a portion of computer component usage that is not specific to the threads to each of the types of transactions.

15. The one or more processor readable storage devices of claim 14, said code for programming said processor to determine the utilization of each of the one or more computer components by the threads comprises code for programming said processor to:

determine a utilization of the one or more processors for performing the plurality of transactions.

16. The one or more processor readable storage devices of claim 15, wherein said code for programming said processor to determine the utilization of each of the one or more computer components by the threads comprises code for programming said processor to:

determine a time used to process the plurality of transactions by the one or more processors.

17. The one or more processor readable storage devices of claim 14, said code for programming said processor to determine the utilization of each of the one or more computer components by the threads comprises code for programming said processor to:

determine a utilization of a memory system for performing the plurality of transactions.

18. The one or more processor readable storage devices of claim 17, wherein said code for programming said processor to determine the utilization of each of the one or more computer components by the threads comprises code for programming said processor to:

determine a size of data stored to the memory system due to processing the plurality of transactions.

19. The one or more processor readable storage devices of claim 18, wherein the data stored to the memory system is one or more allocated objects.

20. The one or more processor readable storage devices of claim 17, the memory system comprising random access memory.

21. The one or more processor readable storage devices of claim 17, wherein the memory system comprises a hard disk device.

22. The one or more processor readable storage devices of claim 14, said code for programming said processor to determine the utilization of each of the one or more computer components by the threads comprises code for programming said processor to:

determine a utilization of network bandwidth for the plurality of transactions.

23. The one or more processor readable storage devices of claim 14, wherein said code for programming said processor to determine the utilization of each of the one or more computer components by the threads comprises code for programming said processor to:

monitor the data communicated through a network socket allocated for the plurality of transactions.

24. The one or more processor readable storage devices of claim 14, wherein the one or more computer components comprise a central processing unit and random access memory residing on one machine.

25. A computer implemented method, comprising:

monitoring a first type of transaction of a plurality of types of transactions that are executed by a system using one or more resources;

receiving application runtime data regarding the one or more resources based on the monitoring, the data comprising transaction-specific application runtime data that is directly attributable to resource usage of the first type of transaction and non-transaction specific data that is indirectly attributable to resource usage of the plurality of types of transactions; and

determining an amount of each of the one or more resources used for the first type of transaction based on the transaction-specific application runtime data and the non-transaction specific data, the determining comprises attributing a portion of non-transaction specific usage of the one or more resources by the plurality of types of transactions to the first type of transaction.

26. The computer implemented method of claim 25, further comprising:

receiving a URL request, the first type of transaction performed in response to the URL request.

27. The computer implemented method of claim 25, wherein the resources comprise a CPU, RAM memory, hard disk memory and network socket connections.

28. The computer implemented method of claim 25, wherein the plurality of types of transactions are performed as part of a computer process, further comprising determining an amount of each of the one or more resources used for the plurality of transactions, comprising:

determining a first amount of each resource used during the computer process to directly perform each of the plurality of types of transactions, and

determining a second amount of each resource used during the computer process which is not directly attributed to individual ones of the plurality of types of transactions.

29. The computer implemented method of claim 28, wherein the second amount of each resource used is attributed to the first type of transaction based on the percentage of resource usage for the first type of transaction with respect to resource usage for the plurality of types of transactions.

30. The computer implemented method of claim 28, further comprising determining the resources needed for an expected occurrence of the types of transactions based on the amount of each resource used for each of the types of transactions.

31. The computer implemented method of claim 25, further comprising:

determining the number of resources needed to meet an expected capacity based on the amount of each of the one or more resources used for the first type of transaction.

32. One or more non-transitory processor readable storage devices having processor readable code embodied on said processor readable storage devices, said processor readable code for programming one or more processors to:

monitor two or more types of transactions in response to requests to perform the two or more types of transactions using one or more resources; and

determine a first usage level of each of the one or more resources for a first type of transaction of the two or more types of transactions based on the monitoring, the first usage level is based on resource utilization that is directly attributable to the first type of transaction; and

determine a second usage level for the one or more resources to attribute to the first type of transaction based on the monitoring, the second usage level indirectly attributable to the first type of transaction, the determining the second usage level comprises apportioning resource usage of the two or more types of transactions that is indirectly attributable to the first type of transaction to the first type of transaction.

33. The one or more processor readable storage devices of claim 32, the requests received as a URL request, wherein the code for programming one or more processors to determine a first usage comprises code for programming said one or more processors to associate the first type of transaction with a uniform resource locator.

34. The one or more processor readable storage devices of claim 32, wherein the second usage is apportioned to the first type of transaction by a load percentage of the first type of transaction with respect to the total load of the two or more type of transactions that use the one or more resources.

35. The one or more processor readable storage devices of claim 34, wherein the type of transactions are performed as part of a computer process.

36. The one or more processor readable storage devices of claim 35, wherein the computer process is a Java Virtual Machine (JVM) process.

37. The one or more processor readable storage devices of claim 32, wherein said code for programming said one or more processors to determine a second usage level comprises code for programming said one or more processors to:

determine the total usage level of the one or more resources by the two or more types of transactions during a period of time as part of a computer process.

38. The one or more processor readable storage devices of claim 37, wherein said code for programming said one or more processors to determine a second usage level comprises code for programming said one or more processors to:

determine the total resource use during the period of time with respect to the computer process.

39. An apparatus for processing data, comprising:

one or more resources;

a processor coupled to the one or more resources, the processor configured to;

monitor a plurality of types of transactions that are executed by an application using the one or more resources;

determine, for each of the types of transactions, a first usage of each of the one or more resources that is directly attributable to each of the plurality of types of transactions based on the monitoring; and

associate a second usage of each of the one or more resources with one or more of the plurality of types of transactions based on the monitoring, the second usage associated with non-transaction specific usage of the one or more resources, the associating comprises attributing a portion of the non-transaction specific usage of the one or more resources to each of the one or more types of transactions.

40. The apparatus of claim 39, wherein the one or more resources comprise at least one of:

a hard disk, a storage device, or a communication interface.