Today's cutting edge network hardware features extremely low latency and high bandwidth transactions for higher-level communication substrates. The Cray XC/XE family of network fabrics, also known as Cray Aries/Gemini respectively, supports such high-performance remote memory access operations (RMA) and a plethora of transaction modes to optimize communication via lower-level interfaces such as uGNI and DMAPP. However, enabling efficient one-sided communication for higher-level substrates is difficult due to barriers presented by the programming model itself, as well as miscellaneous synchronization bottlenecks at the runtime layers. We present an efficient programming model based on a distributed memory allocator for RMA and a communication substrate based on readers and writers for inter-node message passing and RMA operations. We try to maximize performance by introducing a scalable RMA event notification scheme and synchronization protocols that fully leverage Aries/Gemini fabric. Micro-benchmark results demonstrate that our library outperforms Cray MPI-3.0-based RMA one-sided operations by 1.5X and up to 6X in certain cases and is comparable or improves upon performance on others.
机构:
Chinese Acad Sci, Inst Comp Technol, Res Ctr Adv Comp Syst, Beijing, Peoples R ChinaChinese Acad Sci, Inst Comp Technol, Res Ctr Adv Comp Syst, Beijing, Peoples R China
Zhang, Ke
Chang, Yisong
论文数: 0引用数: 0
h-index: 0
机构:
Chinese Acad Sci, Inst Comp Technol, Res Ctr Adv Comp Syst, Beijing, Peoples R ChinaChinese Acad Sci, Inst Comp Technol, Res Ctr Adv Comp Syst, Beijing, Peoples R China
Chang, Yisong
Zhang, Lixin
论文数: 0引用数: 0
h-index: 0
机构:
Chinese Acad Sci, Inst Comp Technol, Res Ctr Adv Comp Syst, Beijing, Peoples R ChinaChinese Acad Sci, Inst Comp Technol, Res Ctr Adv Comp Syst, Beijing, Peoples R China
Zhang, Lixin
Chen, Mingyu
论文数: 0引用数: 0
h-index: 0
机构:
Chinese Acad Sci, Inst Comp Technol, Res Ctr Adv Comp Syst, Beijing, Peoples R ChinaChinese Acad Sci, Inst Comp Technol, Res Ctr Adv Comp Syst, Beijing, Peoples R China
Chen, Mingyu
Yu, Lei
论文数: 0引用数: 0
h-index: 0
机构:
Chinese Acad Sci, Inst Comp Technol, Res Ctr Adv Comp Syst, Beijing, Peoples R ChinaChinese Acad Sci, Inst Comp Technol, Res Ctr Adv Comp Syst, Beijing, Peoples R China
Yu, Lei
Xu, Zhiwei
论文数: 0引用数: 0
h-index: 0
机构:
Chinese Acad Sci, Inst Comp Technol, Res Ctr Adv Comp Syst, Beijing, Peoples R ChinaChinese Acad Sci, Inst Comp Technol, Res Ctr Adv Comp Syst, Beijing, Peoples R China
Xu, Zhiwei
[J].
2016 16TH IEEE/ACM INTERNATIONAL SYMPOSIUM ON CLUSTER, CLOUD AND GRID COMPUTING (CCGRID),
2016,
: 560
-
569