Managing risk in resource over-committed systems

摘要

Risk associated with over-committing shared resources is determined. In response to receiving a request to provision a new workload, a candidate mix of virtual machines is selected from plurality of virtual machines already running on a cloud infrastructure. A utilization profile is then created for an aggregate behavior of the candidate mix of virtual machines and a new virtual machine running the new workload. A risk inherent in over-commitment if the new workload is grouped with the candidate mix of virtual machines is determined, and whether that risk is acceptable. If the risk is acceptable, the new workload is provisioned by over-committing the candidate mix of virtual machines with the new virtual machine running on the cloud infrastructure.

主权项

1. A computer-implemented method for provisioning a virtual machine onto shared physical resources, the method comprising:
identifying a set of workload characteristics for a candidate mix of virtual machines; determining a probability that an aggregate resource utilization by the virtual machine and the candidate mix of virtual machines exceeds a predefined set threshold for resource utilization over a time horizon, wherein the step of determining the probability that the aggregate resource utilization by the candidate mix of virtual machines exceeds the predefined set threshold for resource utilization over the time horizon further comprises: creating a profile of a set of time samples for the aggregate resource utilization by the candidate mix; determining a sample mean and sample standard deviation from the created profile; setting the predefined set threshold to a one-sided upper tolerance limit of the created profile; determining a coverage factor for the created profile; and determining an upper bound on the probability of exceeding the set threshold; and provisioning the virtual machine along with the candidate mix of virtual machines to a physical machine.