OpenFAM: Programming disaggregated memory

High performance computing (HPC) clusters are increasingly handling workloads where working data sets cannot be easily partitioned or are too large to fit into local node memory. In order to enable HPC workloads to access memory external to the node, HPE has defined a programming API (OpenFAM) for d...

Full description

Saved in:
Bibliographic Details
Published in:Concurrency and computation 2023-12, Vol.35 (28), p.n/a
Main Authors: Singhal, Sharad, Crasta, Clarete Riana, Abdulla K, Mashood, Barmawer, Faizan, Bhat, Gautham, Rao, Ramya Ahobala, P N, Soumya, Rajak, Rishi Kesh K
Format: Article
Language:English
Subjects:
Clusters
Computer memory
contexts
disaggregated memory
fabric attached memory
high performance computing
interleaved RDMA
multi‐threading
programming API
Workload
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:High performance computing (HPC) clusters are increasingly handling workloads where working data sets cannot be easily partitioned or are too large to fit into local node memory. In order to enable HPC workloads to access memory external to the node, HPE has defined a programming API (OpenFAM) for developing applications that use large‐scale disaggregated memory. In this paper we describe an open‐source reference implementation of OpenFAM that can be used on scale‐up machines, traditional HPC clusters, as well as emerging disaggregated memory architectures. We demonstrate the efficiency of the implementation using micro‐benchmarks on InfiniBand and Slingshot‐based clusters.
ISSN:1532-0626
1532-0634
DOI:10.1002/cpe.7910