Parallel-aware, dedicated job co-scheduling within/across symmetric multiprocessing nodes
US7810093B2 · kind B2 · utility
Assignee
Inventors
Key dates
| Filing date | Nov 15, 2004 |
| Grant date | Oct 5, 2010 |
| Priority date | — |
| Expiry date | Apr 30, 2029 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06F9/4825
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
In a parallel computing environment comprising a network of SMP nodes each having at least one processor, a parallel-aware co-scheduling method and system for improving the performance and scalability of a dedicated parallel job having synchronizing collective operations. The method and system uses a global co-scheduler and an operating system kernel dispatcher adapted to coordinate interfering system and daemon activities on a node and across nodes to promote intra-node and inter-node overlap of said interfering system and daemon activities as well as intra-node and inter-node overlap of said synchronizing collective operations. In this manner, the impact of random short-lived interruptions, such as timer-decrement processing and periodic daemon activity, on synchronizing collective operations is minimized on large processor-count SPMD bulk-synchronous programming styles.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.