TY - GEN
T1 - FULT
T2 - 47th International Conference on Parallel Processing, ICPP 2018
AU - Dang, Hoang Vu
AU - Snir, Marc
N1 - Funding Information:
This research was supported by the Exascale Computing Project (17-SC-20-SC), a joint project of the U.S. Department of Energy’s Office of Science and National Nuclear Security Administration. This work used the Extreme Science and Engineering Discovery Environment [29], supported by NSF grant numbers ACI-1548562, through allocation TG-CCR130058.
Publisher Copyright:
© 2018 Association for Computing Machinery.
PY - 2018/8/13
Y1 - 2018/8/13
N2 - This paper describes FULT, a user-level thread scheduling system that uses bit-vectors to represent runnable threads. This system is aimed at efficient support of event driven task scheduling. We show a significant reduction in the cost of signal and wait primitives, high scalability, and similar performance for task spawning and other operations, compared conventional task schedulers that use work queues.
AB - This paper describes FULT, a user-level thread scheduling system that uses bit-vectors to represent runnable threads. This system is aimed at efficient support of event driven task scheduling. We show a significant reduction in the cost of signal and wait primitives, high scalability, and similar performance for task spawning and other operations, compared conventional task schedulers that use work queues.
UR - http://www.scopus.com/inward/record.url?scp=85054877652&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85054877652&partnerID=8YFLogxK
U2 - 10.1145/3225058.3225115
DO - 10.1145/3225058.3225115
M3 - Conference contribution
AN - SCOPUS:85054877652
SN - 9781450365109
T3 - ACM International Conference Proceeding Series
BT - Proceedings of the 47th International Conference on Parallel Processing, ICPP 2018
PB - Association for Computing Machinery
Y2 - 14 August 2018 through 16 August 2018
ER -