Memory-aware scheduling for NUMA architectures

专利名称：Memory-aware scheduling for NUMA

architectures

发明人：Alin Jula,Jan Karstens,Alexander Froemmgen申请号：US12981143申请日：20101229公开号：US08635626B2公开日：20140121

专利附图：

摘要：A topology reader may determine a topology of a Non-Uniform MemoryAccess (NUMA) architecture including a number of, and connections between, a pluralityof sockets, each socket including one or more cores and at least one memory configured

to execute a plurality of threads of a software application. A core list generator maygenerate, for each designated core of the NUMA architecture, and based on thetopology, a proximity list listing non-designated cores in an order corresponding to aproximity of the non-designated cores to the designated core. A core selector maydetermine, at a target core and during the execution of the plurality of threads, that thetarget core is executing an insufficient number of the plurality of threads, and may selecta source core at the target core, according to the proximity list associated therewith, forsubsequent transfer of a transferred thread from the selected source core to the targetcore for execution thereon.

申请人：Alin Jula,Jan Karstens,Alexander Froemmgen

地址：Mountain View CA US,Mannheim DE,Mannheim DE

国籍：US,DE,DE

代理机构：Brake Hughes Bellermann LLP

更多信息请下载全文后查看