XaaS: Acceleration as a Service to Enable Productive High-Performance Cloud Computing

被引:1
作者
Hoefler, Torsten [1 ]
Copik, Marcin [1 ]
Beckman, Pete [2 ]
Jones, Andrew [3 ]
Foster, Ian [4 ]
Parashar, Manish [5 ]
Reed, Daniel [6 ]
Troyer, Matthias [3 ]
Schulthess, Thomas [7 ]
Ernst, Daniel [8 ]
Dongarra, Jack [9 ]
机构
[1] Swiss Fed Inst Technol, CH-8093 Zurich, Switzerland
[2] Argonne Natl Lab, Lemont, IL 60439 USA
[3] Microsoft, Redmond, WA 98052 USA
[4] Argonne Natl Lab, Data Sci & Learning Div, Lemont, IL 60439 USA
[5] Utah Univ, Kalhert Sch Comp, Salt Lake City, UT 84112 USA
[6] Univ Utah, Comp Sci & Elect & Comp Engn, Salt Lake City, UT 84117 USA
[7] Swiss Natl Supercomp Ctr, CH-6900 Lugano, Switzerland
[8] Nvidia, Santa Clara, CA 95051 USA
[9] Univ Tennessee, Knoxville, TN 37996 USA
关键词
Cloud computing; Containers; Computational modeling; Computer architecture; Meteorology; Ecosystems; Resource management;
D O I
10.1109/MCSE.2024.3382154
中图分类号
TP39 [计算机的应用];
学科分类号
081203 ; 0835 ;
摘要
High-performance computing (HPC) and the cloud have evolved independently, specializing their innovations into performance or productivity. Acceleration as a Service (XaaS) is a recipe to empower both fields with a shared execution platform that provides transparent access to computing resources, regardless of the underlying cloud or HPC service provider. Bridging HPC and cloud advancements, XaaS presents a unified architecture built on performance-portable containers. Our converged model concentrates on low-overhead, high-performance communication and computing, targeting resource-intensive workloads from climate simulations to machine learning. XaaS lifts the restricted allocation model of Function as a Service (FaaS), allowing users to benefit from the flexibility and efficient resource utilization of serverless computing while supporting long-running and performance-sensitive workloads from HPC.
引用
收藏
页码:40 / 51
页数:12
相关论文
共 17 条
[1]   Enabling High-Performance Computing as a Service [J].
AbdelBaky, Moustafa ;
Parashar, Manish ;
Kim, Hyunjoo ;
Jordan, Kirk E. ;
Sachdeva, Vipin ;
Sexton, James ;
Jamjoom, Hani ;
Shae, Zon-Yin ;
Pencheva, Gergina ;
Tavakoli, Reza ;
Wheeler, Mary F. .
COMPUTER, 2012, 45 (10) :72-80
[2]  
Armbrust M., 2009, Above the clouds: A berkeley view of cloud computing
[3]   Stateful Dataflow Multigraphs: A Data-Centric Model for Performance Portability on Heterogeneous Architectures [J].
Ben-Nun, Tal ;
Licht, Johannes de Fine ;
Ziogas, Alexandros N. ;
Schneider, Timo ;
Hoefler, Torsten .
PROCEEDINGS OF SC19: THE INTERNATIONAL CONFERENCE FOR HIGH PERFORMANCE COMPUTING, NETWORKING, STORAGE AND ANALYSIS, 2019,
[4]   Sarus: Highly Scalable Docker Containers for HPC Systems [J].
Benedicic, Lucas ;
Cruz, Felipe A. ;
Madonna, Alberto ;
Mariotti, Kean .
HIGH PERFORMANCE COMPUTING: ISC HIGH PERFORMANCE 2019 INTERNATIONAL WORKSHOPS, 2020, 11887 :46-60
[5]   Utilising urgent computing to tackle the spread of mosquito-borne diseases [J].
Brown, Nick ;
Nash, Rupert ;
Poletti, Piero ;
Guzzetta, Giorgio ;
Manica, Mattia ;
Zardini, Agnese ;
Flatken, Markus ;
Vidal, Jules ;
Gueunet, Charles ;
Belikov, Evgenij ;
Tierny, Julien ;
Podobas, Artur ;
Chien, Wei Der ;
Markidis, Stefano ;
Gerndt, Andreas .
PROCEEDINGS OF URGENTHPC 2021: THE THIRD INTERNATIONAL WORKSHOP ON HPC FOR URGENT DECISION MAKING, 2021, :36-44
[6]   rFaaS: Enabling High Performance Serverless with RDMA and Leases [J].
Copik, Martin ;
Taranov, Konstanfin ;
Calotoiu, Alexandru ;
Hoefler, Torsten .
2023 IEEE INTERNATIONAL PARALLEL AND DISTRIBUTED PROCESSING SYMPOSIUM, IPDPS, 2023, :897-907
[7]  
DONGARRA JJ, 1998, SOFTW ENVIRONM TOOL, P1
[8]  
Foster I., 2017, Cloud computing for science and engineering
[10]  
Foster I, 2008, GCE: 2008 GRID COMPUTING ENVIRONMENTS WORKSHOP, P60