Exploring memory hierarchy and network topology for runtime AMR data sharing across scientific applications

Wenzhao Zhang, Houjun Tang, Stephen Ranshous, Surendra Byna, Daniel F. Martin, Kesheng Wu, Bin Dong, Scott Klasky, Nagiza F. Samatova

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

2 Scopus citations

Abstract

Runtime data sharing across applications is of great importance for avoiding high I/O overhead for scientific data analytics. Sharing data on a staging space running on a set of dedicated compute nodes is faster than writing data to a slow disk-based parallel file system (PFS) and then reading it back for post-processing. Originally, the staging space has been purely based on main memory (DRAM), and thus was several orders of magnitude faster than the PFS approach. However, storing all the data produced by large-scale simulations on DRAM is impractical. Moving data from memory to SSD-based burst buffers is a potential approach to address this issue. However, SSDs are about one order of magnitude slower than DRAM. To optimize data access performance over the staging space, methods such as prefetching data from SSDs according to detected spatial access patterns and distributing data across the network topology have been explored. Although these methods work well for uniform mesh data, which they were designed for, they are not well suited for adaptive mesh refinement (AMR) data. Two maȩjor issues must be addressed before constructing such a memory hierarchy and topology-aware runtime AMR data sharing framework: (1) spatial access pattern detection and prefetching for AMR data; (2) AMR data distribution across the network topology at runtime. We propose a framework that addresses these challenges and demonstrate its effectiveness with extensive experiments on AMR data. Our results show the framework's spatial access pattern detection and prefetching methods demonstrate about 26% performance improvement for client analytical processes. Moreover, the framework's topology-aware data placement can improve overall data access performance by up to 18%.

Original languageEnglish
Title of host publicationProceedings - 2016 IEEE International Conference on Big Data, Big Data 2016
EditorsRonay Ak, George Karypis, Yinglong Xia, Xiaohua Tony Hu, Philip S. Yu, James Joshi, Lyle Ungar, Ling Liu, Aki-Hiro Sato, Toyotaro Suzumura, Sudarsan Rachuri, Rama Govindaraju, Weijia Xu
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1359-1366
Number of pages8
ISBN (Electronic)9781467390040
DOIs
StatePublished - 2016
Event4th IEEE International Conference on Big Data, Big Data 2016 - Washington, United States
Duration: Dec 5 2016Dec 8 2016

Publication series

NameProceedings - 2016 IEEE International Conference on Big Data, Big Data 2016

Conference

Conference4th IEEE International Conference on Big Data, Big Data 2016
Country/TerritoryUnited States
CityWashington
Period12/5/1612/8/16

Fingerprint

Dive into the research topics of 'Exploring memory hierarchy and network topology for runtime AMR data sharing across scientific applications'. Together they form a unique fingerprint.

Cite this