×

Management of multiple scale out workloads

  • US 9,515,905 B1
  • Filed: 10/31/2008
  • Issued: 12/06/2016
  • Est. Priority Date: 10/31/2008
  • Status: Active Grant
First Claim
Patent Images

1. A computer-implemented process comprising:

  • monitoring nodes executing instances of workloads to gather respective performance metrics, said workloads including plural scale-out workloads each having plural scale-out instances,for each of said scale-out workloads, aggregating by a computer the performance metrics across instances of the scale-out workload to obtain an aggregate performance metric representative of an aggregate performance of the scale-out workload so as to obtain respective aggregate performance metrics for said scale-out workloads; and

    based at least in part on said aggregate performance metrics, allocating by a computer an additional node to at least one scale-out workload by selecting a node on which a logical-server template for said scale-out workload has been instantiated or by instantiating a logical-server template said additional node, said additional node being a physical server, a hard or soft partition of a physical server, or a virtual machine.

View all claims
  • 2 Assignments
Timeline View
Assignment View
    ×
    ×