US 7,734,676 B2Granted

Method for controlling the number of servers in a hierarchical resource environment

AssigneeInternational Business Machines Corporation

InventorsGreg M. Dritschler, Guenter Vater, Robert Vaupel, Peter B. Yocom

Life of this patent

Estimating the term…

Loading patent…

Abstract

The invention relates to the control of servers which process client work requests in a computer system on the basis of resource consumption. Each server contains multiple server instances (also called “execution units”) which execute different client work requests in parallel. A workload manager determines the total number of server containers and server instances in order to achieve the goals of the work requests. The number of server instances started in each server container depends on the resource consumption of the server instances in each container and on the resource constraints, service goals and service goal achievements of the work units to be executed. At predetermined intervals during the execution of the work units the server instances are sampled to check whether they are active or inactive. Dependent on the number of active server instances the number of server address spaces and server instances is repeatedly adjusted to achieve an improved utilization of the available virtual storage and an optimization of the system performance in the execution of the application programs.

Claims11

Dependent claims

A method for controlling the number of server instances in a computer system controlled by an operating system having server regions that are managed by a workload manager to provide resources for achieving service goals of work units received from application programs, said server regions including a number of server containers each containing a plurality of server instances operating in parallel to execute said work units received from said application programs, comprising the steps of:

(a) sampling the server instances in a server container at execution time to obtain sample data representing resource usage of server instances that are active during a predetermined sampling interval, said sample data indicating resource contention among said server instances;
(b) evaluating the sample data to determine a current resource consumption of the computer system in executing the work units;
(c) (1) calculating an optimal number of server instances per server container from the current resource consumption of the computer system to execute the work units, said step comprising providing threshold values for current resource consumption and using said threshold values to permit or prevent an adjustment of the number of server instances in at least one of the server container, said step comprising restricting the number of server instances in each server container on the basis of restrictions on the total number of server instances permitted for all server containers executing work units for one or more server classes;
(2) calculating an optimal total number of server instances, executing in parallel in each of the server containers, to execute the work units;
(3) calculating a number of server containers from the optimal total number of server instances and from the optimal number of server instances per server container;
and
(d) providing feedback to the workload manager based upon the calculated number of server containers and the calculated number of server instances per server container, thereby to cause said workload manager to adjust the number of server containers and the number of server instances in at least one server container;
and
(e) repeating steps (a) to (d) at predetermined time intervals during execution of the work units.

6 more elements

A computer program product for controlling the number of server instances in a computer system controlled by an operating system having server regions that are managed by a workload manager to provide resources for achieving service goals of work units received from application programs, said server regions including a number of server containers each containing a plurality of server instances operating in parallel to execute said work units received from said application programs, the computer program product comprising program code means stored on a non-transitory computer readable medium that runs on a computer system to perform the following steps:

(a) sampling the server instances in a server container at execution time to obtain sample data representing resource usage of server instances that are active during a predetermined sampling interval, said sample data indicating resource contention among said server instances;
(b) evaluating the sample data to determine a current resource consumption of the computer system in executing the work units;
(c) (1) calculating an optimal number of server instances per server container from the current resource consumption of the computer system, to execute the work units, said step comprising providing threshold values for current resource consumption and using threshold values to permit or prevent an adjustment of the number of server instances in at least one of the server containers, said step comprising restricting the number of server instances in each server container on the basis of restrictions on the total number of server instances permitted for all server containers executing work units for one or more server classes;
(2) calculating an optimal total number of server instances, executing in parallel in each of the server containers, to execute the work units;
(3) calculating a number of server containers from the optimal total number of server instances and from the optimal number of server instances per server container;
and
(d) providing feedback to the workload manager based upon the calculated number of server containers and the calculated number of server instances per server container, thereby to cause said workload manager to adjust the number of server containers and the number of server instances in at least one server;
and
(e) repeating steps (a) to (d) at predetermined time intervals during execution of the work units.

6 more elements

Apparatus for controlling the number of server instances in a computer system controlled by an operating system having server regions that are managed by a workload manager to provide resources for achieving service goals of work units received from application programs, said server regions including a number of server containers each containing a plurality of server instances operating in parallel to execute said work units received from said application programs, comprising:

a hardware processor;
(a) sampling the server instances in a server container at execution time to obtain sample data representing resource usage of server instances that are active during a predetermined sampling interval, said sample data indicating resource contention among said server instances;
(b) evaluating the sample data to determine a current resource consumption of the computer system in executing the work units;
(c) (1) calculating, via the hardware processor an optimal number of server instances per server container from the current resource consumption of the computer system to execute the work units, said step comprising providing threshold values for current resource consumption and using said threshold values to permit or prevent of the number of server instances in at least one of the server containers, said step comprising restricting number of server instances in each server container on the basis of restriction on the total number of server instances permitted for all server containers executing work units for one or more service class;
(2) calculating an optimal total number of server instances, executing in parallel in each of the server containers, to execute the work units;
(3) calculating a number of server containers from the optimal total number of server instances and from the optimal number of server instances per server container;
and
(d) providing feedback to the workload manager based upon the calculated number of server containers and the calculated number of server instances per server container, thereby to cause said workload manager to adjust the number of server containers and the number of server instances in at least one server;
and
(e) repeating steps (a) to (d) at predetermined time intervals during execution of the work units.

7 more elements

Dependencies are inferred from the claim text.