International Business Machines Corporation

Robert Vaupel

Guenter Vater

Pradeep Parameshwaran

Wilhelm Mild

Omar-Qais Noorshams

Nicolas X Nardelli

Mariia Zrianina

Meik Neubauer

Verena Schlinkmann

Peter Bergersen Yocom

Donna N Dillenberger

Stephen John Heisig

Ingolf Salm

Stefan Wirag

Verena Steinmetz

Greg M Dritschler

Georg W Bosch

A method, computer program product, and system includes a processor(s) in a first computing environment obtaining, from a computer system in the first environment, a system message indicating an issue in the computer system. The processor(s) evaluate the message to determine details of the issue and determine recommended action(s) to remedy the issue. The processor(s) generate a notification comprising the action(s), which include calls to portions of the computer system and the notification comprises respective selection options to trigger the processor(s) to execute each of the one or more recommended actions in the first environment. The processor(s) transmit the notification to a computing node in a second computing environment to communicate to a computing device in that environment. The processor(s) receive a response indicating selection of a respective selection option. The processor(s) translate the option into a recommended action and issue calls in the recommended action to the computer system.

Dynamically accessing and configuring secured systems

Mechanisms for distributing workload items to a plurality of dispatchers are provided. Each dispatcher is associated with a different computing system of a plurality of computing systems and workload items comprise workload items of a plurality of different workload types. A capacity value for each combination of workload type and computing system is obtained. For each combination of workload type and computing system, a queue length of a dispatcher associated with the corresponding computing system is obtained. For each combination of workload type and computing system, a dispatcher's relative share of incoming workloads is computed used on the queue length for the dispatcher associated with the computing system. In addition, incoming workload items are routed to a dispatcher, in the plurality of dispatchers, based on the calculated dispatcher's relative share for the dispatcher.

Calculating a dispatcher's relative share based on relative queue length and capacity value of a plurality of workload types and computing systems combinations

An adaptive mechanism is provided that learns the response time characteristics of a workload by measuring the response times of end user transactions, classifies response times into buckets, and dynamically adjusts the response time distribution as response time characteristics of the workload change. The adaptive mechanism maintains the actual distribution across changes and, thus, helps the end user to understand changes of workload behavior that take place over a longer period of time. The mechanism is stable enough to suppress spikes and returns a constant view of workload behavior, which is required for long term, performance analysis and capacity planning. The mechanism distinguishes between an initial learning phase of establishing the distribution and one or multiple reaction periods. The reaction periods can be for example a fast reaction period for strong fluctuations of the workload behavior and a slow reaction period for small deviations.

Method for adaptively determining response time distribution of transactional workloads by adjusting the midpoint of response time distribution

Apparatus and computer program product for adaptively determining response time distribution of transactional workloads

Mechanisms for distributing workload items to a plurality of dispatchers are provided. Each dispatcher is associated with a different computing system of a plurality of computing systems and workload items comprise workload items of a plurality of different workload types. A capacity value for each combination of workload type and computing system is obtained. For each combination of workload type and computing system, a queue length of a dispatcher associated with the corresponding computing system is obtained. For each combination of workload type and computing system, a dispatcher's relative share of incoming workloads is computed based on the queue length for the dispatcher associated with the computing system. In addition, incoming workload items are routed to a dispatcher, in the plurality of dispatchers, based on the calculated dispatcher's relative share for the dispatcher.

Routing workloads based on relative queue lengths of dispatchers

Determining an equivalent capacity (ECP) of a computing system comprising multi-type operational resources. The multi-type operational resources comprises at least one general type of resources and at least one specialized type of resources Parameters characteristic of the performance of the system is determined. Assignment of work units to the various resources subject to pre-defined constraints is simulated. Utilization of said general type of resources of the computing system when executing the work units is calculated.

Managing a workload in a cluster of computing systems with multi-type operational resources

The present invention provides a new method and a new system for controlling the capacity usage of a logically partitioned data processing system comprising physical data processing resources that are divided into multiple logical partitions, wherein the capacity of the logical partitions is monitored to control the consumption of the physical data processing resources. The new method and the system are characterized in that at least one group of logical partitions is defined that is managed towards a common capacity group limit.

Method for controlling the capacity usage of a logically partitioned data processing system

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.

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

A computer program product stored on computer storage media includes instructions for managing a workload in a computing system. The product including instructions for collecting resource and state data for a plurality of service class periods, calculating a long term execution velocity, if sufficient data exists, then determining a goal velocity using the data, otherwise, selecting a default value for the goal velocity, associating the goal velocity with the respective service class period, repeating the calculating, determining, selecting and associating for each service class period in the plurality of service class periods, and adjust the resource access to each of the service class periods according to the associated goal velocities.

Autonomous management of system throughput

In a heterogeneuos work environment a method and apparatus for managing isochronous streaming servers is disclosed. This includes a mechanism for allocating processor resource to streaming servers at admission control time with respect to the business importance of work they will be running. This mechanism includes a provision to preempt currently running streaming work of lesser business importance in order to allocate its share of the processor to the new higher importance work. Also included is a mechanism to resume the server when a running stream terminated and processor becomes available. In the case of the variable bit rate multiple peak problem a mechanism is presented to preempt running streaming work at the lowest business importance so that higher priority work can continue to run unaffected. The mechanism to resume paused streams at the termination of a stream also restarts this work. In this way streaming work is never degraded so as to become unintelligible due to processor constraints but either runs correctly or is paused/terminated. Also presented is a method of combining a queuing model, admission control, process management and preemption/resumption services to allow a generic streaming service or isochronous process to be integrated into a heterogeneous work environment on a particular platform.

Growing community of inventors

Rottenburg, Germany

Robert Vaupel