发明名称 Method and apparatus for determining to-be-migrated task based on cache awareness
摘要 A method and an apparatus for determining a to-be-migrated task based on cache awareness in a computing system having multiple processor cores is disclosed. In the method, the computing system determines a source processor core and a destination processor core according to a load of each processor core. Through respectively monitoring the number of cache misses of each task and the number of executed instructions of each task in the source processor core and the destination processor core, the computing system obtain an average cache miss per kilo instructions of the source processor core and an average cache miss per kilo instructions of the destination processor core. Then, the computing system determines, according to the obtained average cache miss per kilo instructions of the source processor core and the destination processor core, a task to be migrated from the source processor core to the destination processor core.
申请公布号 US9483321(B2) 申请公布日期 2016.11.01
申请号 US201514676195 申请日期 2015.04.01
申请人 Huawei Technologies Co., Ltd. 发明人 Xu Yuanchao;Fan Dongrui;Zhang Hao;Ye Xiaochun
分类号 G06F9/46;G06F13/00;G06F9/50;G06F9/48 主分类号 G06F9/46
代理机构 Conley Rose, P.C. 代理人 Conley Rose, P.C. ;Rodolph Grant;Beaulieu Nicholas K.
主权项 1. A method for determining a to-be-migrated task based on cache awareness in a computing system having multiple processor cores, comprising; obtaining information about each of the multiple processor cores, wherein the information comprises load information of each processor core; determining a source processor core and a destination processor core according to the load information of each processor core; monitoring the number of cache misses of each task and the number of executed instructions of each task in the source processor core; obtaining a cache miss ratio of each task in the source processor core by dividing the number of cache misses of the each task by the number of executed instructions of the each task; calculating an average cache miss ratio of the source processor core according to cache miss ratios of all tasks in the source processor core; monitoring the number of cache misses of each task and the number of executed instructions of each task in the destination processor core; obtaining a cache miss ratio of each task in the destination processor core by dividing the number of cache misses of the each task by the number of executed instructions of the each task; calculating an average cache miss ratio of the destination processor core according to cache miss ratios of all tasks in the destination processor core; and migrating a task that is with a smallest cache miss ratio and is in the source processor core to the destination processor core when the average cache miss ratio of the source processor core is not less than the average cache miss ratio of the destination processor core; or migrating a task that is with a largest cache miss ratio and is in the source processor core to the destination processor core when the average cache miss ratio of the source processor core is less than the average cache miss ratio of the destination processor core.
地址 Shenzhen CN