摘要 |
<p>A method and system are disclosed for managing resources within a compute environment. The method embodiment of the invention comprises receiving, at a first time, a request from a requestor to consume compute resources in the compute environment, wherein the request is associated with a workload to be submitted at a second time later than the first time. An analysis of compute resources in the compute environment is performed, based on an agreement between the requestor and the compute environment for a guaranteed quality of service for compute resources. Based on the analysis and in compliance with the agreement, compute resources in the compute environment are reserved to yield a reservation. The workload submitted by the requestor is received and processed by the compute resources in the compute environment.</p> |