2021 IEEE/ACM 11th Workshop on Fault Tolerance for HPC at eXtreme Scale (FTXS) 2021
DOI: 10.1109/ftxs54580.2021.00006
|View full text |Cite
|
Sign up to set email alerts
|

Assessing the Use Cases of Persistent Memory in High-Performance Scientific Computing

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1

Citation Types

0
4
0

Year Published

2022
2022
2023
2023

Publication Types

Select...
3
1

Relationship

1
3

Authors

Journals

citations
Cited by 4 publications
(4 citation statements)
references
References 28 publications
0
4
0
Order By: Relevance
“…Fridman et al [8] present an evaluation of local PMFSs (e.g., ext4-DAX and SplitFS [11]) for writing and reading diagnostics of scientific applications, as well as for performing C/R of these applications using transparent or explicit checkpointing. While these local PMFSs are good choices for single-node workloads, scientific computing applications typically require a distributed PMFS operating on multiple nodes.…”
Section: Persistent Memory File Systems (Pmfss)mentioning
confidence: 99%
See 2 more Smart Citations
“…Fridman et al [8] present an evaluation of local PMFSs (e.g., ext4-DAX and SplitFS [11]) for writing and reading diagnostics of scientific applications, as well as for performing C/R of these applications using transparent or explicit checkpointing. While these local PMFSs are good choices for single-node workloads, scientific computing applications typically require a distributed PMFS operating on multiple nodes.…”
Section: Persistent Memory File Systems (Pmfss)mentioning
confidence: 99%
“…NVRAM's ability to retain data even after node failures, opens new possibilities for high-performance computing, most notably, as a medium for data persistence upon node or process failure and consequent recovery. Previous investigation [8] of the ways in which Intel Optane™ DCPMM can be used for supporting the recoverability of scientific applications has established that it enables enhanced checkpoint-restart support, either explicitly or transparently. This paper explores how NVRAM can be exploited to support ESR, while reducing the cost of redundancy, specifically in the context of PCG solvers.…”
Section: Introductionmentioning
confidence: 99%
See 1 more Smart Citation
“…As a result, academic and industrial researchers have proposed various middleware and libraries, such as databases [7,19,31], hash tables [16,17,27], and tree-structured indexes [11,18,33,34]. Although Intel Optane, the first persistent memory product, is no longer in production, vendors have continued to develop technologies such as Storage Class Memory (SCM) and Compute Express Link (CXL) that will reach persistent memory [6].…”
Section: Introductionmentioning
confidence: 99%