Building portable thread schedulers for hierarchical multiprocessors: the bubblesched framework
作者： Samuel Thibault , Raymond Namyst , Pierre-André Wacrenier
DOI: 10.1007/978-3-540-74466-5_6
关键词:
摘要: Exploiting full computational power of current more and hierarchical multiprocessor machines requires a very careful distribution threads data among the underlying non-uniform architecture. Unfortunately, most operating systems only provide poor scheduling API that does not allow applications to transmit valuable hints system. In previous paper [1], we showed using bubble-based thread scheduler can significantly improve applications' performance in portable way. However, since multithreaded have various requirements, there is no universal could meet all these needs. this paper, present framework allows experts implement experiment with customized schedulers. It provides powerful for dynamically distributing bubbles machine high-level, portable, efficient Several examples show how then develop, debug tune their own bubble
springer.com
arxiv.org
archives-ouvertes.fr
sci-hub.st 参考文章(8)
Pascal Hénon, Pierre Ramet, Jean Roman, PaStiX: A Parallel Sparse Direct Solver Based on a Static Scheduling for Mixed 1D/2D Block Distributions international parallel and distributed processing symposium. ,vol. 1800, pp. 519- 527 ,(2000) , 10.1007/3-540-45591-4_70
Vincent Danjean, Raymond Namyst, Pierre-André Wacrenier, An Efficient Multi-level Trace Toolkit for Multi-threaded Applications Euro-Par 2005 Parallel Processing. ,vol. 3648, pp. 166- 175 ,(2005) , 10.1007/11549468_21
Xipen Shen, Yaoqing Gao, Chen Ding, Roch Archambault, Lightweight reference affinity analysis Proceedings of the 19th annual international conference on Supercomputing - ICS '05. pp. 131- 140 ,(2005) , 10.1145/1088149.1088167
Jaydeep Marathe, Frank Mueller, Hardware profile-guided automatic page placement for ccNUMA systems acm sigplan symposium on principles and practice of parallel programming. pp. 90- 99 ,(2006) , 10.1145/1122971.1122987
D. Durand, T. Montaut, L. Kervella, W. Jalby, Impact of memory contention on dynamic scheduling on NUMA multiprocessors IEEE Transactions on Parallel and Distributed Systems. ,vol. 7, pp. 1201- 1214 ,(1996) , 10.1109/71.544359
M. D. Durand, T. Montaut, L. Kervella, W. Jalby, Impact of Memory Contention on Dynamic Scheduling on Numa Multiprocessors international conference on parallel processing. ,vol. 1, pp. 258- 262 ,(1993) , 10.1109/ICPP.1993.101
Alexandra Fedorova, Margo I. Seltzer, Operating system scheduling for chip multithreaded processors Harvard University. ,(2006)
Samuel Thibault, A Flexible Thread Scheduler for Hierarchical Multiprocessor Machines ieee international conference on high performance computing data and analytics. ,(2005)