Abstract:
A method, information processing system, and computer readable storage medium are provided for dynamically managing accelerator resources. A first set of hardware accelerator resources (122) is initially assigned to a first information processing system (104), and a second set of hardware accelerator resources (124) is initially assigned to a second information processing system (106). Jobs running on the first and second information processing systems (116, 118) are monitored (132). When one of the jobs fails to satisfy a goal, at least one hardware accelerator resource in the second set of hardware accelerator resources from the second information processing system are dynamically reassigned (138) to the first information processing system.
Abstract:
The allocation of shareable resources of a computing environment is dynamically adjusted to balance the workload of that environment. Workload is managed across two or more partitions of a plurality of partitions of the computing environment, which are preferably configured as groups of partitions. At least one group of the computing environment includes a plurality of partitions of the computing environment. Shareable resources are assigned to the partitions of the group and are managed as a group. The managing includes dynamically adjusting allocation of a shareable resource of at least one partition of the two or more partitions in order to balance workload goals of the two or more partitions. One example of this is managing central processing unit (CPU) resources 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. The adjusting includes modifying processor weights associated with the partitions.
Abstract:
A method, information processing system, and computer readable storage medium are provided for dynamically managing accelerator resources. A first set of hardware accelerator resources (122) is initially assigned to a first information processing system (104), and a second set of hardware accelerator resources (124) is initially assigned to a second information processing system (106). Jobs running on the first and second information processing systems (116, 118) are monitored (132). When one of the jobs fails to satisfy a goal, at least one hardware accelerator resource in the second set of hardware accelerator resources from the second information processing system are dynamically reassigned (138) to the first information processing system.
Abstract:
A partitioned processing system capable of supporting diverse operating syst em partitions is disclosed wherein throughput information is passed from a partition to a partition resource manager. The throughput information is use d to create resource balancing directives for the partitioned resource. The processing system includes at least a first partition and a second partition . A partition resource manager is provided for receiving information about throughput from the second partition and determining resource balancing directives. A communicator communicates the resource balancing directives fr om the partition manager to a kernel in the second partition which allocates resources to the second partition according to the resource balancing directives received from the partition manager.
Abstract:
A method, information processing system, and computer readable storage medium are provided for dynamically managing accelerator resources. A first set of hardware accelerator resources (122) is initially assigned to a first information processing system (104), and a second set of hardware accelerator resources (124) is initially assigned to a second information processing system (106). Jobs running on the first and second information processing systems (116, 118) are monitored (132). When one of the jobs fails to satisfy a goal, at least one hardware accelerator resource in the second set of hardware accelerator resources from the second information processing system are dynamically reassigned (138) to the first information processing system.
Abstract:
Un método para administrar carga de trabajo en un entorno de ordenador, comprendiendo dicho método: determinar que se ha ajustado una asignación de recursos de CPU a una partición de dicho entorno de ordenador; y ajustar dinámicamente dicha asignación a través de dicha partición y otra partición de dicho entorno de ordenador; en el que dicha partición comprende una o más clases de servicio de trabajo, y en el que dicha determinación está caracterizada porque comprende: proyectar un efecto en una clase de servicio seleccionada de dichas una o más clases de servicio de aumentar un peso de procesador asociado con dicha partición (600); proyectar un efecto en una o más clases de servicio de dicha otra partición de disminuir un peso de procesador de dicha otra partición (604), y determinar si el beneficio a dicha clase de servicio seleccionada anula un impacto negativo para el trabajo de dicha otra partición (606), en que el ajuste ha de hacerse cuando el beneficio anule el impacto negativo.
Abstract:
The allocation of shareable resources of a computing environment is dynamically adjusted to balance the workload of that environment. Workload i s managed across two or more partitions of a plurality of partitions of the computing environment, which are preferably configured as groups of partitions. At least one group of the computing environment includes a plurality of partitions of the computing environment. Shareable resources ar e assigned to the partitions of the group and are managed as a group. The managing includes dynamically adjusting allocation of a shareable resource o f at least one partition of the two or more partitions in order to balance workload goals of the two or more partitions. One example of this is managin g central processing unit (CPU) resources 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. The adjusting includ es modifying processor weights associated with the partitions.