International Business Machines Corporation

Catherine Krueger Eilert

Peter Bergersen Yocom

Bernard R Pierce

Jeffrey David Aman

Gary M King

John Elliott Arwe

Harry M Yudenfriend

Jeffrey Mark Nick

Jeffrey Paul Kubala

Casper A Scalzi

Richard J Schmalz

David Andrew Booz

Donald H Gibson

David Vincent Bostjancic

Kenneth G Rubsam

Eugene S Schulze

Gregory M Dritschler

A technique is disclosed for managing a workload distributed across multiple data processing systems to enhance shared resource access to meet a common performance standard. The technique includes on at least one system, measuring performance of the work units on the system to create local performance data, and on at least some of the systems sending the local performance data to at least one other system of the multiple data processing systems. The method further includes on at least one of the systems, receiving the performance data from the sending systems to create remote performance data, and adjusting at least one control parameter for accessing shared resources in response to the local and remote performance data to modify the performance of the work units distributed across the data processing systems to achieve the common performance standard. A dynamic resource clustering process is also employed to enhance the shared resource management.

Workload management method to enhance shared resource access in a multisystem environment

Managing processor resources in a non-dedicated computer system. An amount of a processor resource is allocated to a real-time application of the computer system. The amount does not exceed a limit chosen for a group of real-time applications, wherein the group includes the real-time application being allocated the resource. A selected amount of the processor resource remains available to execute other types of applications and work on the system. During processing of the real-time application, use of the processor resource does not exceed a chosen maximum value, thereby ensuring the processor resource is not monopolized by the real-time application and allowing other types of work to be processed on the system.

System of managing processor resources in a non-dedicated computer system

Disclosed is a method and computer program device for dynamically managing the assignment of alias addresses to base addresses referencing an input/output (I/O) device, such as a direct access storage device (DASD). Two distinct methods are disclosed. In one method, alias addresses are assigned based on the performance of the I/O devices. In this method, alias addresses are assigned to highly utilized devices, as indicated by device performance data, in order to maximize the efficient utilization of I/O device resources. In a second method, workload management principles are utilized to assign alias addresses. In this method, a correlation is made between each I/O device and the service classes utilizing each device. As in the first method, performance data is generated for each I/O device. Alias addresses are assigned to I/O devices experiencing queue delays as indicated by their performance data, if the device is associated with a service class that has failed to meet one or more processing goals. These methods may operate on a single host, or in a multi-host environment. The methods may be operated individually, or concurrently. Methods are disclosed to manage contention between concurrently operating assignment methods, and between multiple hosts concurrently operating one or more assignment methods.

Dynamic management of addresses to an input/output (I/O) device

Central processing unit (CPU) resources are managed within a computing environment. When the allocation of CPU resources to a partition of the computing environment is to be adjusted, the allocation is adjusted dynamically. This dynamic adjustment is across at least two partitions of the computing environment. The adjusting includes modifying processor weights associated with the partitions.

Method, system and program products for managing central processing unit resources of a computing environment

A method for managing a workload distributed across data processing systems in accordance with a common processor consumption standard, including the steps of measuring the processor consumption of the work units to create local processor consumption data; sending the local processor consumption data to at least one other system; receiving processor consumption data from at least one other system to create remote processor consumption data; and adjusting at least one of the system control parameters to modify the remote processor consumption and the local processor consumption of the work units to achieve the common processor consumption standard is provided. Also provided is an apparatus according to the method.

Multi-system resource capping

Managing processor resources in a non-dedicated computer system

A method and apparatus for controlling the number of servers in a multisystem cluster. Incoming work requests are organized into service classes, each of which has a queue serviced by servers across the cluster. Each service class has defined for it a local performance index for each particular system of the cluster and a multisystem performance index for the cluster as a whole. Each system selects one service class as a donor class for donating system resources and another service class as a receiver class for receiving system resources, based upon how well the service classes are meeting their goals. Each system then determines the resource bottleneck causing the receiver class to miss its goals. If the resource bottleneck is the number of servers, each system determines whether and how many servers should be added to the receiver class, based upon whether the positive effect of adding such servers on the performance index for the receiver class outweighs the negative effect of adding such servers on the performance measure for the donor class. If a system determines that servers should be added to the receiver class, it then determines the system in the cluster to which the servers should be added, based upon the effect on other work on that system. To make this latter determination, each system first determines whether another system has enough idle capacity and, if so, lets that system add servers. If no system has sufficient idle capacity, each system then determines whether the local donor class will miss its goals if servers are started locally. It not, the servers are started on the local system. Otherwise, each system determines where the donor class will be hurt the least and acts accordingly. To ensure the availability of a server capable of processing each of the work requests in the queue, each system determines whether there is a work request in the queue with an affinity only to a subset of the cluster that does not have servers for the queue and, if so, starts a server for the queue on a system in the subset to which the work request has an affinity.

Method and apparatus for controlling the number of servers in a multisystem cluster

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.

Method and apparatus for controlling the number of servers in a

A system within a multisystem environment is selected to run a real-time application. The selection process looks for a system in which the amount of a processor resource for delivering a real-time data stream of the real-time application has the least impact on other work processing on the selected system. The real-time application is allocated an amount of a processor resource that does not exceed a limit chosen for a group of one or more real-time applications. A selected amount of the processor resource remains available for at least one non-real-time application of the multisystem environment.

System for managing processor resources in a multisystem environment in

A system within a multisystem environment is selected to run a real-time application. The selection process looks for a system in which the amount of a processor resource for delivering a real-time data stream of the real-time application has the least impact on other work proseccing on the selected system. The real-time application is allocated an amount of a processor resource that does not exceed a limit chosen for a group of one or more real-time applications. A selected amount of the processor resource remains available for at least one non-real-time application of the multisystem environment.

Managing processor resources in a multisystem environment in order to

An apparatus for managing a workload distributed across data processing systems in accordance with a common performance standard, which includes a means for measuring the performance of the work units to create local performance data; a means for sending said local performance data to at least one other system; a means for receiving performance data from at least one other system to create remote performance data; and a means responsive to said local and remote performance data for adjusting at least one of the system control parameters to modify the performance of the work units on the system to achieve the common performance standard, is disclosed. Also disclosed is a method for managing workload as represented by the apparatus.

Apparatus and method for managing a distributed data processing system

A workload manager creates an in storage representation of a set of performance goals, each goal associated with a class of clients (e.g., client transactions) in a client/server data processing system. A set of servers, providing service to the clients, are managed to bring the clients into conformity with the class performance goals by: calculating performance indexes for each class to determine the target class(es) which are farthest behind their class performance goals; analyzing the relationship among servers and client classes to determine which servers serve which classes; determining which resource(s) are impacting the service provided to the key servers (that is, those on which the target class(es) are most heavily reliant), and projecting the effect of making more of these resources available to those servers; and, finally, making the changes to those resources which are projected to most favorably indirectly affect the performance of the target class(es).

Apparatus and method for managing a server workload according to client

A workload manager creates goal control data, defining two or more classes of system work units, in response to specification of goals of two or more goal types for the classes, and specification of importance values for each of the goal types. A system resource manager causes the goals to be met by periodically sampling work unit status; calculating a performance index for each class; selecting a receiver class to receive improved service based on the relative performance indexes and goal importance; a system bottleneck impacting achievement of goal by the receiver class is identified; and one or more system control data elements are identified and adjusted to cause the goal to be met for the receiver class.

Apparatus and method for managing a data processing system workload

Paging data is used to manage address space working set size, system multiprogramming level, and job mix, to improve system productive CPU utilization. System paging characteristics are monitored to determine when to perform management functions. When needed, address space paging data is collected for monitored address spaces, and target working set sizes set for managed address spaces. If necessary, address spaces are 'forced' to exhibit characteristic paging patterns. Reallocation of working set pages among active address spaces, and swap in-swap out actions, are assessed, and impacts on productive CPU utilization is estimated.

Growing community of inventors

Wappingers Falls, NY, United States of America

Catherine Krueger Eilert