2016 IEEE International Parallel and Distributed Processing Symposium (IPDPS) 2016
DOI: 10.1109/ipdps.2016.89
|View full text |Cite
|
Sign up to set email alerts
|

X: A Comprehensive Analytic Model for Parallel Machines

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1

Citation Types

0
3
0

Year Published

2016
2016
2023
2023

Publication Types

Select...
6
2

Relationship

2
6

Authors

Journals

citations
Cited by 17 publications
(3 citation statements)
references
References 15 publications
0
3
0
Order By: Relevance
“…This is especially desired when porting traditional CPU-based HPC applications onto the new GPU-based exascale systems, such as Summit [6], Sierra [7] and Perlmutter [37]. As part of the community effort, we are planning to pursue these research directions in our future work with our past experience on GPU analytic modeling [38], [39], [40] and performance optimization [41], [42], [43], [44], [45], [46], [47], [48], [49], [50], [51].…”
Section: Discussionmentioning
confidence: 99%
“…This is especially desired when porting traditional CPU-based HPC applications onto the new GPU-based exascale systems, such as Summit [6], Sierra [7] and Perlmutter [37]. As part of the community effort, we are planning to pursue these research directions in our future work with our past experience on GPU analytic modeling [38], [39], [40] and performance optimization [41], [42], [43], [44], [45], [46], [47], [48], [49], [50], [51].…”
Section: Discussionmentioning
confidence: 99%
“…As we mentioned in Section 3.2, path-sampling methodology achieves the minimal sampling overhead and consumes significantly less memory compared to other methods. Building a parallel uncertain graph processing framework over GPUs is still challenging [33,34,56]. This subsection focuses on the implementation of challenging phases, which are parallel path identification (Figure 4(a)) and sampling (Figure 4(b)).…”
Section: Gpu-based Design For Asynchronous Path Identification and Sa...mentioning
confidence: 99%
“…Another roofline derived model is the work proposed by Li et al in their "X model" [52]. It is based on an older proposed visual model by the same authors called "Transit" model and combines rooflines of memory bandwidth and compute throughput.…”
Section: Roofline Derived Modelsmentioning
confidence: 99%