Method and apparatus for capacity consumption profiling in a client/server environment

• US 6,725,454 B1
• Filed: 08/21/2000
• Issued: 04/20/2004
• Est. Priority Date: 08/21/2000
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Dilley, Reginald James, Oustalet, John Augustin III, Nagel, Lynn Nowers, Schmidt-Dannert, Ralf H.
• Primary Examiner(s): Banankhah, Majid A.
• Application Number:
  US09/643,075
• Time in Patent Office: 1,338 Days
• Field of Search: 709/100, 709/101, 709/102, 709/103, 709/104, 709/105, 709/220, 709/221, 709/222, 709/223, 709/203, 709/226, 718/100, 718/101, 718/102, 718/103, 718/104, 717/127, 717/128, 717/130
• US Class Current: 718/100
• CPC Class Codes: G06F 11/3409 for performance assessment
• CPC Class Codes: G06F 11/3419 by assessing time
• CPC Class Codes: G06F 11/3452 Performance evaluation by statistical analysis
• CPC Class Codes: G06F 11/3495 for systems

Abstract

A system and method for identifying capacity consumption profiles in a client/server computer system, which include such applications as enterprise resource planning customer relationship planning and supply change management. These profiles may be used for improving the ability to forecast those computer requirements necessary to accommodate a user'"'"'s future data processing requirements. Additionally, these profiles allow for the categorizing of different types of computer usage for a specific client/server solution in a particular industry, which can provide sizing guidance to new users of these client/server solutions. In the present method, selected available statistics and accounting information are consolidated by a context aware process followed by statistical data analysis. After the desired types of profiles are selected with their available client/server identification, the context of the work effort that is being measured is tracked. This allows otherwise undefined activities to be identified in terms of the other associated activities being performed, and then allocated to the correct consumption profile. This is followed by statistical analysis and reduction of the intermediate by a context sensitive reduction process.

Claims

1. A method for monitoring the central processing unit resource capacity consumption in an information handling system having at least two computers in communication over a network, comprising the steps of:

creating a set of profiles to identify selected activities that use CPU resources;

receiving sampled activity CPU usage data from a software application program running on said CPU;

determining if said sampled activity data matches one of said set of profiles;

if no match is determined, assigning said activity data to the most recently identified profile;

receiving actual activity CPU usage data from an operating system regarding execution of said software application program;

determining the variability of said sampled activity data to said actual activity data;

correlating and correcting said received data; and

assigning said CPU capacity consumption to each profile in said set of profiles according to said corrected received data.

2. The method of claim 1, wherein said step of determining the variability comprises the step of applying a least squares analysis of said sampled activity data and said actual activity data.

3. A method for monitoring a (CPU) central processing unit resource capacity consumption in an information handling system having at least two computers in communication over a network, comprising the steps of:

creating a set of profiles to identify selected activities that use CPU resources;

executing a software application program;

during execution, collecting both (a) a stream of sampled activity CPU usage data from the software application program running on said CPU, and (b) a stream of actual activity CPU usage data from an operating system regarding execution of said software application program;

assigning said stream of sampled activity CPU usage data to one of the profiles; and

determining the variability of the stream of sampled activity CPU usage data with the stream of actual activity CPU usage data.

4. The method of claim 3, comprising the further step of time stamping both the stream of sampled activity CPU usage data and the stream of actual activity CPU usage data after they are generated so that step of determining the variability can be performed.