Program-level performance tuning

• US 8,161,462 B2
• Filed: 01/11/2008
• Issued: 04/17/2012
• Est. Priority Date: 09/19/2003
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Herbeck, David Gerald, Hubka, David E., Masbruch, Mark Donald, Perkins, Mark Anthony, Peterson, Joseph Harold, Rackham, DeVaughn Lawrence, Smith, Richard Michael
• Primary Examiner(s): Rutten, James D
• Application Number:
  US12/008,521
• Publication Number:
  US 20080134181A1
• Time in Patent Office: 1,558 Days
• Field of Search: None
• US Class Current: 717/127
• CPC Class Codes: G06F 11/3409 for performance assessment
• CPC Class Codes: G06F 11/3433 for load management allocation of a server based on load conditions G06F9/505; load rebalancing G06F9/5083; redistributing the load in a network by a load balancer H04L67/1029
• CPC Class Codes: G06F 2201/81 Threshold
• CPC Class Codes: G06F 2201/87 Monitoring of transactions

Abstract

A method, apparatus, system, and signal-bearing medium that in an embodiment request a program or programs to tune themselves to run faster or slower if a service class is not meeting its performance goal. In an embodiment, the program is repeatedly requested to incrementally tune itself until the performance goal is met or until no further improvement occurs. In various embodiments, the programs to be requested to tune themselves are selected based on whether the programs are bottlenecks for the service class, whether the programs do the majority of work for the service class, whether the programs easily meet their own performance goals, or whether the programs are low priority. In this way, the programs may be performance tuned in a way that is more effective and less intrusive than by adjusting global, system-level resource allocations.

Claims

1. A method comprising:

receiving a plurality of notifications from a plurality of self-tuning programs, wherein each of the plurality of notifications indicates that the respective self-tuning program is capable of tuning performance of the respective self-tuning program;

determining that a service class fails to meet a performance goal;

sending a tuning request to a first self-tuning program of the plurality of self-tuning programs, wherein the tuning request instructs the first self-tuning program to change the performance of the first self-tuning program; and

selecting the first self-tuning program from among the plurality of self-tuning programs, wherein the deciding that the first self-tuning program is associated with a majority of work in the service class, wherein the deciding further comprises deciding that the first self-tuning program is associated with transactions within a time period whose aggregate response time is greater than half of a response time for all transactions in the service class during the time period.

2. The method of claim 1, wherein the tuning request instructs the first-tuning program to incrementally self-tune.

3. The method of claim 1, further comprising:

determining whether the service class meets the performance goal after the sending.

4. The method of claim 3, further comprising:

if the service class does not meet the performance goal after the sending, determining whether performance of the service class has improved;

if the performance of the service class has improved, repeating the sending so long as the service class does not meet the performance goal and the performance of the service class improves; and

if the performance of the service class has not improved, requesting the first self-tuning program to reset to a previous tuned state.

5. A method comprising:

receiving a plurality of notifications from a plurality of respective self-tuning programs, wherein each of the plurality of notifications indicates that the respective self-tuning program supports tuning of its respective own performance;

determining that a service class failed to meet a performance goal;

selecting a first self-tuning program from among the plurality of self-tuning programs, wherein the selecting further comprises deciding that the first self-tuning program is associated with a majority of work in the service class, wherein the deciding further comprises deciding that the first self-tuning program is associated with transactions within a time period whose aggregate response time is greater than half of a response time for all transactions in the service class during the time period;

requesting the first self-tuning program to incrementally self-tune the performance of the first self-tuning program; and

determining whether the requesting caused performance of the service class to improve.

6. The method of claim 5, wherein the requesting the first self-tuning program to incrementally self-tune further comprises:

requesting the first self-tuning program to increase performance of the first-self tuning program if the first self-tuning program is a performance bottleneck for the service class.

7. The method of claim 5, wherein the requesting further comprises:

requesting the first self-tuning program to self-tune a resource associated with the first self-tuning program.

8. The method of claim 5, further comprising:

if the requesting did not cause the performance of the service class to improve, selecting a second self-tuning program from among the plurality of self-tuning programs and requesting the second self-tuning program to incrementally self-tune.