Method and apparatus for controlling the number of servers in a client/server system

• US 5,974,462 A
• Filed: 03/28/1997
• Issued: 10/26/1999
• Est. Priority Date: 03/28/1997
• Status: Expired due to Term

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Bostjancic, David V., Dritschler, Gregory M., Aman, Jeffrey D., Booz, David A., Yocom, Peter B., Arwe, John E., Eilert, Catherine K.
• Primary Examiner(s): Rinehart, Mark H.
• Application Number:
  US08/828,440
• Time in Patent Office: 942 Days
• Field of Search: 395/200.33, 395/672, 395/673, 395/674, 395/675, 709/203, 709/225, 709/224, 709/223, 709/102, 709/103, 709/104, 709/105
• US Class Current: 709/225
• CPC Class Codes: H04L 67/1001 for accessing one among a plurality of replicated servers
• CPC Class Codes: H04L 67/1008 based on parameters of servers, e.g. available memory or workload monitoring of computer activity G06F11/30
• CPC Class Codes: H04L 67/1031 Controlling of the operation of servers by a load balancer, e.g. adding or removing servers that serve requests

Abstract

A method and apparatus for controlling the number of servers in an information handling system in which incoming work requests belonging to a first service class are placed in a queue for processing by one or more servers. The system also has units of work assigned to a second service class that acts as a donor of system resources. In accordance with the invention, a performance measure is defined for the first service class as well as for the second service class. Before adding servers to the first service class, there is determined not only the positive effect on the performance measure for the first service class, but also the negative effect on the performance measure for the second service class. Servers are added to the first service class only if the positive effect on the performance measure for the first service class outweighs the negative effect on the performance measure for the second service class.

Claims

1. In an information handling system in which incoming work requests belonging to a first service class are placed in a queue for processing by one or more servers requiring other system resources, said system also having units of work assigned to one or more other service classes, a method of controlling the number of said servers, comprising the steps of:

defining a performance measure for each of said first and said other service classes;

determining the positive effect on the performance measure for said first service class of adding a predetermined number of servers to said first service class;

selecting one of said other service classes as a donor class from which to obtain the other system resources necessary to add said predetermined number of servers to said first service class;

determining the negative effect on the performance measure for said donor service class of obtaining from said donor class said other system resources necessary to add said predetermined number of servers to said first service class; and

adding said predetermined number of servers to said first service class only if the positive effect on the performance measure for said first service class outweighs the negative effect on the performance measure for said donor service class.

2. The method of claim 1 in which said performance measure is based at least in part upon response time.

3. The method of claim 1 in which said performance measure is based at least in part upon execution velocity.

7. The method of claim 1 in which said other system resources comprise storage.

9. The method of claim 1 in which said step of determining the positive effect on the performance measure for said first service class of adding a predetermined number of servers to said first service class comprises the steps of:

determining a current queue delay for said first service class;

determining a projected queue delay for said first service class with said predetermined number of servers added to said first service class; and

comparing said projected queue delay with said current queue delay.

4. In an information handling system in which incoming work requests belonging to a first service class are placed in a queue for processing by one or more servers requiring other system resources, said system also having units of work assigned to one or more other service classes, apparatus for controlling the number of said servers, comprising:

means for defining a performance measure for each of said first and said other service classes;

means for determining the positive effect on the performance measure for said first service class of adding a predetermined number of servers to said first service class;

means for selecting one of said other service classes as a donor class from which to obtain the other system resources necessary to add said predetermined number of servers to said first service class;

means for determining the negative effect on the performance measure for said donor service class of obtaining from said donor class said other system resources necessary to add said predetermined number of servers to said first service class; and

means for adding said predetermined number of servers to said first service class only if the positive effect on the performance measure for said first service class outweighs the negative effect on the performance measure for said donor service class.

8. The apparatus of claim 4 in which said other system resources comprise storage.

10. The apparatus of claim 4 in which said means for determining the positive effect on the performance measure for said first service class of adding a predetermined number of servers to said first service class comprises:

means for determining a current queue delay for said first service class;

means for determining a projected queue delay for said first service class with said predetermined number of servers added to said first service class; and

means for comparing said projected queue delay with said current queue delay.

5. A program storage device readable by a machine, tangibly embodying a program of instructions executable by the machine to perform method steps for controlling the number of servers assigned to a first service class in an information handling system in which incoming work requests belonging to said first service class are placed in a queue for processing by one or more of said servers, said servers requiring other system resources, said system also having units of work assigned to one or more other service classes, said method steps comprising:

defining a performance measure for each of said first and said other service classes;

determining the positive effect on the performance measure for said first service class of adding a predetermined number of servers to said first service class;

selecting one of said other service classes as a donor class from which to obtain the other system resources necessary to add said predetermined number of servers to said first service class;

determining the negative effect on the performance measure for said donor service class of obtaining from said donor class said other system resources necessary to add said predetermined number of servers to said first service class; and

adding said predetermined number of servers to said first service class only if the positive effect on the performance measure for said first service class outweighs the negative effect on the performance measure for said donor service class.

6. The program storage device of claim 5 in which said other system resources comprise storage.

11. The program storage device of claim 5 in which said step of determining the positive effect on the performance measure for said first service class of adding a predetermined number of servers to said first service class comprises the steps of:

determining a current queue delay for said first service class;

determining a projected queue delay for said first service class with said predetermined number of servers added to said first service class; and

comparing said projected queue delay with said current queue delay.