International Business Machines Corporation

Thomas Lumpp

Juergen P Schneider

Oliver Benke

Andrew Neil Trossman

Isabell Schwertle

Ming Chen

Wolfgang Schaeberle

Boas Betzler

Eberhard Pasch

Reinhard Theodor Buendgen

David B Petersen

Ana Claudia Biazetti

Jeffrey S Lucash

Chun-Shi Chang

Markus Mueller

John K Alex

Rong-Sheng Lee

Utz Bacher

Peter Richard Badovinatz

Gregory D Laib

Sven Sterbling

Martin Reitz

Bernd Jostmeyer

Walter Schueppen

Jeurgen Schneider

Wolfgang Kuechlin

Carsten Sinz

Joachim Gönnheimer

A system may include a memory and a processor in communication with the memory. The processor may be configured to perform operations. The operations may include loading environment rules and available workloads and retrieving workload data for the available workloads, execution environment data of an execution environment, and historic workload data. The operations may include selecting a workload for the execution environment and determining execution settings for the workload. The operations may include submitting the workload to an execution queue for execution in the execution environment according to the execution settings.

Central randomized scheduler for hypothesis-based workloads

A method for dynamically creating of a high availability policy based upon entries in a configuration management database (CMDB) and a best practices template can be provided. The method can include directing resource discovery in a CMDB for a selected application to discover components and dependencies of the selected application. The method also can include mapping the components and dependencies discovered for the selected application in the CMDB to a best practices template specifying a parameterized high availability topology for the selected application. Finally, the method can include storing the best practices template with mapped components and dependencies as a high availability policy for the selected application.

Dynamic high availability policy creation based upon entries in a configuration management database (CMDB) and a best practices template

A method for dynamically creating a high availability policy based upon entries in a configuration management database (CMDB) and a best practices template can be provided. The method can include directing resource discovery in a CMDB for a selected application to discover components and dependencies of the selected application. The method also can include mapping the components and dependencies discovered for the selected application in the CMDB to a best practices template specifying a parameterized high availability topology for the selected application. Finally, the method can include storing the best practices template with mapped components and dependencies as a high availability policy for the selected application.

A running of an application resource of a first site of at least two sites is stopped from accessing a storage device of the first site. A running of a replication resource is brought to a stop. The replication resource is configured to control a replication of data between the storage device on the first site on which data from the running application resource on the first site are stored to the storage device on the second site. A replication request is submitted to the automation engine for changing a replication direction to run from the second site to the first site. The application resource is started on the second site only after changing the replication direction to start from the storage device on the second site instead of from the storage device of the first site.

Method for controlling changes of replication directions in a multi-site disaster recovery environment for high available application

An autonomic computing system and method determine that a desired end state cannot be reached, determine that an acceptable sub-state can be reached using at least one of priority ratings, conditional relationship specifications, and alternative relationship specifications, and place the computing system in an acceptable sub-state. The priority ratings can contain an attribute assigned to a policy definition that determines a sequence for applying the policy definition. The attribute can be 'mandatory', a numerical value, or 'not required'. The conditional relationship specifications have policy definitions that are applied when the state of a specified resource meets a predetermined requirement. An alternative relationship specification has policy definitions and/or conditional relationship specifications that are applied when the state of a specified resource does not meet a predetermined requirement.

Policy driven autonomic computing-programmatic policy definitions

A method, apparatus, and computer instructions are provided for a common cluster model for configuring, managing, and operating different clustering technologies in a data center. The common cluster model supports peer cluster domains and management server cluster domains. Each cluster domain may have one or more cluster nodes. For each cluster domains, one or more cluster resources may be defined. These resources may depend on one another and may be grouped into a resource group. A set of cluster domain and cluster resources logical device operations are provided to configure, manage, and operate cluster domains and its associated resources.

Common cluster model for configuring, managing, and operating different clustering technologies in a data center

Provided are techniques for determining a recovery time for a resource in a heterogeneous computing environment comprising interdependent resources. A graph for the resource representing all sequence dependencies and all group relations are created. The recovery time may be a cumulative startup time or a cumulative shutdown time of the resource considering interdependencies of the resource to other resources. The recovery time for all support resources having sequence dependencies with the resource is calculated and each node representing the support resources are removed from the graph. Then the recovery time for all member resources left in the graph that have group relations with the resource is calculated per a group type of the resource. The recovery time for the resource is a sum of the recovery time of all support resources, the recovery time of all member resources, and a unit recovery time of the resource.

Determining recovery time for interdependent resources in heterogeneous computing environment

A mechanism is provided for determining an incident of a resource in a computing environment. An event pertaining to the resource is processed by a system automation module. The event is represented as an associated event data having parameters of a target state, a target state prior to the event, a current state, and a current state prior to the event. First, the target state is compared to the target state prior to the event to assure that the target state is steady. Wherein a determination that the event is an incident cannot be made after comparing the target state and the current state, the system automation module compares the current state to the current state prior to the event. Upon determining that the event is an incident, the event data is marked and stored in a repository.

Determining availability parameters of resource in heterogeneous computing environment

A system and associated method for determining a recovery time for a resource in a heterogeneous computing environment comprising interdependent resources. A graph for the resource representing all sequence dependencies and all group relations are created. The recovery time may be a cumulative startup time or a cumulative shutdown time of the resource considering interdependencies of the resource to other resources. The recovery time for all support resources having sequence dependencies with the resource is calculated and each node representing the support resources are removed from the graph. Then the recovery time for all member resources left in the graph that have group relations with the resource is calculated per a group type of the resource. The recovery time for the resource is a sum of the recovery time of all support resources, the recovery time of all member resources, and a unit recovery time of the resource.

A resource broker is used in managing workloads of a peer-to-peer distributed computing environment. The broker is responsible for distributing to a plurality of clients of the peer-to-peer distributed computing environment tasks of one or more workloads of one or more issuers of the environment. The tasks are processed by the clients using a runtime environment provided by the broker. The runtime environment of a client is generic to the one or more issuers and independent of an operating system of the client.

Employing a resource broker in managing workloads of a peer-to-peer computing environment

A method, apparatus, and computer instructions are provided for expressing high availability (H/A) cluster demand based on probability of breach. When a failover occurs in the H/A cluster, event messages are sent to a provisioning manager server. The mechanism of embodiments of the present invention filters the event messages and translates the events into probability of breach data. The mechanism then updates the data model of the provision manager server and makes a recommendation to the provisioning manager server as to whether reprovisioning of new node should be performed. The provisioning manager server makes the decision and either reprovisions new nodes to the H/A cluster or notifies the administrator of detected poisoning problem.

Method and apparatus for expressing high availability cluster demand based on probability of breach

An apparatus and computer instructions are provided for expressing high availability (H/A) cluster demand based on probability of breach. When a failover occurs in the H/A cluster, event messages are sent to a provisioning manager server. The mechanism of embodiments of the present invention filters the event messages and translates the events into probability of breach data. The mechanism then updates the data model of the provision manager server and makes a recommendation to the provisioning manager server as to whether reprovisioning of new node should be performed. The provisioning manager server makes the decision and either reprovisions new nodes to the H/A cluster or notifies the administrator of detected poisoning problem.

Apparatus expressing high availability cluster demand based on probability of breach

The present invention provides a system and method for the execution of jobs in a distributed computing architecture that uses worker clients which are characterized by a checkpointing mechanism component for generating checkpointing information being assigned to at least one worker client, at least one failover system being assigned to the worker client, a component (failover system selection component) for automatically assigning at least one existing or newly created failover system to the failure system being assigned to a worker client in the case said worker clients fails, wherein the assigned failover system provides all function components in order to take over the execution of the job when said assigned worker client fails, wherein the assigned failover system further includes at least a failover monitor component for detecting failover situations of said assigned worker client.

System and method for execution of a job in a distributed computing architecture

An autonomic computing system and method determine policy definitions () and a set of available actions (); monitor resources distributed within the system; determine if the system is at a desired end state; and modify resource states by sending instructions for the resources to perform available actions. The policy definitions () specify: start order between resources, prioritization between resources, conditional activation of policies, desired end state of resources, and location limitation of resources. The system and method receive status information from available resources, and monitor and modify the system until it reaches the desired end state. The policy definitions () can be determined by specifying a user-defined system end state and resource relationships. The policy definitions () can further harvest implicit relationships between resources, via self-discovery, and underlying relationships among resources.

Growing community of inventors

Reutlingen, Germany

Thomas Lumpp