EXTOLL and Data Movements in Heterogeneous Computing Environments

Holger Fröning ()
Additional contact information
Holger Fröning: Ruprecht-Karls University of Heidelberg, Institute of Computer Engineering

A chapter in Sustained Simulation Performance 2014, 2015, pp 127-139 from Springer

Abstract: Abstract Driven by the end of Dennard scaling, we are observing a fundamental transition to communication-centric computing systems, based on heterogeneous processing units like CPUs, GPUs and others. The vast increase of parallelism and heterogeneity together with the advent of Big Data leads to systems in which communication is more expensive than computation, both in terms of energy and time. This work reviews EXTOLL as a new contender in the HPC interconnect arena, highly optimized for such environments. EXTOLL has been designed from scratch for HPC needs, allowing for slim and efficient software stacks. In the second part we will focus on GGAS, a research project we use to explore data movement optimizations in heterogeneous environments using EXTOLL.

Keywords: Message Passing Interface; Address Space; Execution Model; Communication Path; Incoming Packet (search for similar items in EconPapers)
Date: 2015
References: Add references at CitEc
Citations:

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:sprchp:978-3-319-10626-7_11

Ordering information: This item can be ordered from
http://www.springer.com/9783319106267

DOI: 10.1007/978-3-319-10626-7_11

Access Statistics for this chapter

More chapters in Springer Books from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().