Abstract: A method for managing resources of a computer cluster, wherein automated means allocate to a job at least one resource among several resources from the cluster, the automated means selecting the resource based on at least one historical data relative to previous uses of the resources and/or data relative to the arrangement, temperature, power consumption, bandwidth, or maintenance of the cluster or one or more components thereof, the automated mechanism thus determining the wear of the various resources available and choosing the resources with the lowest wear to perform the job, thereby avoiding the over-utilization of resources, one of the main causes of failures in a computer cluster, and increasing the lifetime of the cluster.

Abstract: The invention relates to a method for organizing tasks, in at least some nodes of a computer cluster, comprising: First, launching two containers on each of said nodes, a standard container and a priority container, next, for all or part of said nodes with two containers, at each node, while a priority task does not occur, assigning one or more available resources of the node to the standard container thereof in order to execute a standard task, the priority container thereof not executing any task, when a priority task occurs, dynamically switching only a portion of the resources from the standard container thereof to the priority container thereof, such that, the priority task is executed in the priority container with the switched portion of the resources, and the standard task continues to be executed, without being halted, in the standard container with the non-switched portion of the resources.

Abstract: A method for managing resources of a computer cluster, wherein automated means allocate to a job at least one resource among several resources from the cluster, the automated means selecting the resource based on at least one historical data relative to previous uses of the resources and/or data relative to the arrangement, temperature, power consumption, bandwidth, or maintenance of the cluster or one or more components thereof, the automated mechanism thus determining the wear of the various resources available and choosing the resources with the lowest wear to perform the job, thereby avoiding the over-utilization of resources, one of the main causes of failures in a computer cluster, and increasing the lifetime of the cluster.

Abstract: The invention relates to a method for organizing tasks, in at least some nodes of a computer cluster, comprising: First, launching two containers on each of said nodes, a standard container and a priority container, next, for all or part of said nodes with two containers, at each node, while a priority task does not occur, assigning one or more available resources of the node to the standard container thereof in order to execute a standard task, the priority container thereof not executing any task, when a priority task occurs, dynamically switching only a portion of the resources from the standard container thereof to the priority container thereof, such that, the priority task is executed in the priority container with the switched portion of the resources, and the standard task continues to be executed, without being halted, in the standard container with the non-switched portion of the resources.

Abstract: A method for determining an optimal frequency (fo) for the execution of a software application on an information processing system. The method includes a first execution of said application at a predetermined frequency (fd), making it possible to determine an overall execution time (Tg) and an overall energy consumption (Eg) and a second execution of said application at the same frequency by means of a measurement tool, making it possible to determine measurements (TMPI*, TIO*, Tg*) on said execution; a step of determining a first law providing an execution time (t(f)) as a function of the frequency (f), and a second law providing an energy consumption (E(f)) as a function of the frequency (f); and a step of determining said optimal frequency (fo) as the frequency optimizing a criterion (C(f)) combining said execution time and said energy consumption, as a function of the frequency.

Abstract: A method for determining an optimal frequency (fo) for the execution of a software application on an information processing system. The method includes a first execution of said application at a predetermined frequency (fd), making it possible to determine an overall execution time (Tg) and an overall energy consumption (Eg) and a second execution of said application at the same frequency by means of a measurement tool, making it possible to determine measurements (TMPI*, TIO*, Tg*) on said execution; a step of determining a first law providing an execution time (t(f)) as a function of the frequency (f), and a second law providing an energy consumption (E(f)) as a function of the frequency (f); and a step of determining said optimal frequency (fo) as the frequency optimizing a criterion (C(f)) combining said execution time and said energy consumption, as a function of the frequency.

Abstract: A system is provided including at least two resource and task managers which are independent of each other; a cluster of shared resources common to these managers; software that runs in the background interfaced with the managers in a manner so as to appropriately distribute the resources of the cluster between the managers on the basis of one or more distribution parameters.