Apparatus and method of dynamically repartitioning a computer system in response to partition workloads
First Claim
1. A computer implemented method of dynamically re-partitioning a partitioned computer system in response to workloads, each partition of the partitioned computer system having a plurality of resources, the computer implemented method comprising:
- creating a workload profile for each partition, the workload profile having a workload and a workload time schedule;
determining, for a partition before the workload of the workload profile is to occur on the partition, that the workload of the workload profile will cause a demand for a resource in the plurality of resources originally assigned to the partition to exceed a maximum threshold;
identifying a second partition including a second plurality of resources, the second plurality of resources including a second resource and a third resource, wherein the second resource and the third resource are similar to one another, and wherein the second resource is operating below a minimum capacity in the second partition;
determining whether the second partition can continue operating without the second resource by using the third resource;
selecting, responsive to determining that the second partition can continue operating without the second resource, the second resource as an additional resource; and
automatically allocating the additional resource to the partition before the workload of the workload profile is executed on the partition.
0 Assignments
0 Petitions
Accused Products
Abstract
A method, system and apparatus for dynamically repartitioning a partitioned computer system in response to workload demands are provided. In one embodiment, a monitoring software is used to monitor workloads on all resources in all the partitions. If a workload on a resource in a partition is determined to exceed a maximum threshold, a similar resource is allocated to the partition. The similar resource is preferentially an unassigned or unallocated resource. However, resources from other partitions may also be used. In another embodiment, a workload schedule is stored in a workload profile. If a scheduled workload in any of the resources of a partition is to exceed a maximum threshold, additional similar resources will be allocated to the partition before the scheduled workload.
9 Citations
24 Claims
-
1. A computer implemented method of dynamically re-partitioning a partitioned computer system in response to workloads, each partition of the partitioned computer system having a plurality of resources, the computer implemented method comprising:
-
creating a workload profile for each partition, the workload profile having a workload and a workload time schedule; determining, for a partition before the workload of the workload profile is to occur on the partition, that the workload of the workload profile will cause a demand for a resource in the plurality of resources originally assigned to the partition to exceed a maximum threshold; identifying a second partition including a second plurality of resources, the second plurality of resources including a second resource and a third resource, wherein the second resource and the third resource are similar to one another, and wherein the second resource is operating below a minimum capacity in the second partition; determining whether the second partition can continue operating without the second resource by using the third resource; selecting, responsive to determining that the second partition can continue operating without the second resource, the second resource as an additional resource; and automatically allocating the additional resource to the partition before the workload of the workload profile is executed on the partition. - View Dependent Claims (2, 3, 4, 5, 6)
-
-
7. A computer program product on a computer readable storage medium for dynamically re-partitioning a partitioned computer system in response to workloads, each partition of the partitioned computer system having a plurality of resources, the computer program product comprising:
-
code for creating a workload profile for each partition, the workload profile having a workload and a workload time schedule; code for determining, for a partition before the workload of the workload profile is to occur on the partition, that the workload of the workload profile will cause a demand for a resource in the plurality of resources originally assigned to the partition to exceed a maximum threshold; code for identifying a second partition including a second plurality of resources, the second plurality of resources including a second resource and a third resource, wherein the second resource and the third resource are similar to one another, and wherein the second resource is operating below a minimum capacity in the second partition; code for determining whether the second partition can continue operating without the second resource by using the third resource; code for selecting, responsive to determining that the second partition can continue operating without the second resource, the second resource as an additional resource; and code for automatically allocating the additional resource to the partition before the workload of the workload profile is executed on the partition. - View Dependent Claims (8, 9, 10, 11, 12)
-
-
13. An apparatus for dynamically re-partitioning a partitioned computer system in response to workloads, each partition of the partitioned computer system having a plurality of resources, the apparatus comprising:
-
a storage device including a storage medium, wherein the storage device stores computer usable program code; and a processor, wherein the processor executes the computer usable program code, and wherein the computer usable program code comprises; computer usable code for creating a workload profile for each partition, the workload profile having a workload and a workload time schedule; computer usable code for determining, for a partition before the workload of the workload profile is to occur on the partition, that the workload of the workload profile will cause a demand for a resource in the plurality of resources originally assigned to the partition to exceed a maximum threshold; computer usable code for identifying a second partition including a second plurality of resources, the second plurality of resources including a second resource and a third resource, wherein the second resource and the third resource are similar to one another, and wherein the second resource is operating below a minimum capacity in the second partition; computer usable code for determining whether the second partition can continue operating without the second resource by using the third resource; computer usable code for selecting, responsive to determining that the second partition can continue operating without the second resource, the second resource as an additional resource; and computer usable code for automatically allocating the additional resource to the partition before the workload of the workload profile is executed on the partition. - View Dependent Claims (14, 15, 16, 17, 18)
-
-
19. A computer system that is partitioned into a plurality of partitions, the computer system having code data for dynamically re-partitioning itself in response to workloads, each partition of the partitioned computer system having a plurality of resources, the computer system comprising:
-
at least one storage device for storing the code data; and at least one processor for processing the code data to create a workload profile for each partition, the workload profile having a workload and a workload time schedule determine, for a partition before the workload of the workload profile is to occur on the partition, that the workload of the workload profile will cause a demand for a resource in the plurality of resources originally assigned to the partition to exceed a maximum threshold; identify a second partition including a second plurality of resources, the second plurality of resources including a second resource and a third resource, wherein the second resource and the third resource are similar to one another, and wherein the second resource is operating below a minimum capacity in the second partition; determine whether the second partition can continue operating without the second resource by using the third resource; select, responsive to determining that the second partition can continue operating without the second resource, the second resource as an additional resource; and automatically allocate the additional resource to the partition before the workload of the workload profile is executed on the partition. - View Dependent Claims (20, 21, 22, 23, 24)
-
Specification