Non-uniform memory access (NUMA) enhancements for shared logical partitions

摘要

In a NUMA-topology computer system that includes multiple nodes and multiple logical partitions, some of which may be dedicated and others of which are shared, NUMA optimizations are enabled in shared logical partitions. This is done by specifying a home node parameter in each virtual processor assigned to a logical partition. When a task is created by an operating system in a shared logical partition, a home node is assigned to the task, and the operating system attempts to assign the task to a virtual processor that has a home node that matches the home node for the task. The partition manager then attempts to assign virtual processors to their corresponding home nodes. If this can be done, NUMA optimizations may be performed without the risk of reducing the performance of the shared logical partition.

主权项

1. A computer system comprising:
a plurality of nodes that each has at least one physical processor coupled to a local memory in a non-uniform memory access (NUMA) topology; a plurality of logical partitions defined on the plurality of nodes, the plurality of logical partitions including a plurality of shared logical partitions that share a pool of common nodes; an operating system in each of the plurality of shared logical partitions that assigns a home node to a task when the task is created, assigns the task to a first virtual processor corresponding to the home node when the first virtual processor is available, and assigns the task to a second virtual processor corresponding to a node other than the home node with a suggestion parameter that specifies the home node when the first virtual processor is not available; and a partition manager that runs each virtual processor on one of the physical processors, the partition manager running the first virtual processor on a physical processor corresponding to the home node when the home node has an available physical processor, the partition manager running the first virtual processor on a physical processor corresponding to a node different than the home node when the home node has no available physical processor.