发明名称 |
Proactive Resource Management for Parallel Work-Stealing Processing Systems |
摘要 |
Various embodiments proactively balance workloads between a plurality of processing units of a multi-processor computing device by making work-stealing determinations based on operating state data. An embodiment method includes obtaining static characteristics data associated with each of a victim processor and one or more of a plurality of processing units that are ready to steal work items from the victim processor (work-ready processors), obtaining dynamic characteristics data for each of the processors, calculating priority values for each of the processors based on the obtained data, and transferring a number of work items assigned to the victim processor to a winning work-ready processor based on the calculated priority values. In some embodiments, the method may include acquiring control over a probabilistic lock for a shared data structure and updating the shared data structure to indicate the number of work items transferred to the winning work-ready processor. |
申请公布号 |
US2017083364(A1) |
申请公布日期 |
2017.03.23 |
申请号 |
US201514862373 |
申请日期 |
2015.09.23 |
申请人 |
QUALCOMM Incorporated |
发明人 |
Zhao Han;Suárez Gracia Dario;Kumar Tushar |
分类号 |
G06F9/48;G06F9/50 |
主分类号 |
G06F9/48 |
代理机构 |
|
代理人 |
|
主权项 |
1. A method implemented by a processor of a multi-processor computing device to proactively balance workloads among a plurality of processing units by making work-stealing determinations based on operating state data, comprising:
obtaining static characteristics data associated with each of a victim processor and one or more of the plurality of processing units that are ready to steal work items from the victim processor (work-ready processors); obtaining dynamic characteristics data including values associated with temperature and power consumption, associated with each of the victim processor and the work-ready processors; calculating priority values for each of the victim processor and the work-ready processors based on the obtained data; and transferring a number of work items assigned to the victim processor to a winning work-ready processor based on the calculated priority values. |
地址 |
San Diego CA US |