LEADER Contents Energy efficiency at exascale
Osman Unsal explains the LEGaTO project, which is aiming to radically reduce the energy consumption of future supercomputers
Laboratory informatics Laboratory in the cloud
The use of cloud resources and software-as-a-service in the laboratory continues to increase. Robert Roe talks to Thermo Fischer about their cloud strategy
Data standards Standardising access to data 14
Sophia Ktori concludes her two-part feature on the creation and use of data standards in the laboratory
17
ACD labs helps users to increase productivity through the use of data standards and the characterising the interpretation of data, writes Sophia Ktori
Modelling and simulation Optimising engineering performance
Optimisation techniques are increasingly valuable to engineers and designers as they enable new methods and increased performance of new components
Resources News
A round-up of the latest news stories in scientific computing Suppliers’ directory
23 26 February/March 2020 l Issue 170
Robert Roe Editor
High performance computing Processor diversity
Robert Roe takes a look at the increasingly competitive world of processors and accelerators for HPC and AI
7 4
Energising economy
10
In this month’s HPC content, there is a theme of energy efficiency. On page 4 we look at the processor technologies available to HPC users. While energy efficiency is always a strong focus for HPC users, the rise of AI is also driving changes in this market. New processing technologies targetted specifically for AI and ML workflows are now available alongside the traditional HPC providers, who are racing to provide the energy efficiency required for exascale. If today’s HPC technology was used
20
in exascale systems, the current power envelope that is being targetted will not be met. However, through the application of a new programming framework, European researchers hope to drastically improve the efficiency of HPC systems while also providing a framework to support CPU, GPU and FPGA resources in a single cluster. On page 7, LEGaTO project coordinator Osman Unsal explains how this new framework could provide an order of magnitude increase in the efficiency of HPC systems. The laboratory informatics section of this issue begins on page 10 with a look at the cloud and SaaS software available to lab users. Based on ELN, LIMS and other laboratory management solutions Thermo Fischer has built a platform that supports both hosted solutions and more traditional in-house cloud deployments. On page 14 Sophia Ktori finishes her
Editorial and administrative team Editor: Robert Roe
robert.roe@
europascience.com Managing editor: Tim Gillett
editor.scw@
europascience.com Specialist reporters: Sophia Ktori, Gemma Church, Design: David Houghton, Zöe Andrews Partnership & Events Executive: Charlie Mitchell
charlie.mitchell@
europascience.com
Advertising team Advertising manager: Mike Nelson
mike.nelson@
europascience.com +44 (0) 1223 221039 Production manager: David Houghton
david.houghton@europascience.com +44 (0) 1223 221034
Corporate team Managing director: Warren Clark
Scientific Computing World is published by Europa Science Ltd, 4 Signet Court, Cambridge, CB5 8LA l ISSN 1744-8026 Tel: +44 (0) 1223 211170 l Fax: +44 (0) 1223 213385 Web:
www.researchinformation.info
Subscriptions: Free registration is available to qualifying individuals. Register online at
www.scientific-computing.com Subscriptions £180 a year for six issues to readers outside registration requirements. Single issue £20. Orders to ESL, SCW Circulation, 4 Signet Court, Swann Road, Cambridge, CB5 8LA, UK. Tel: +44 (0)1223 221 030. Fax: +44 (0)1223 213 385. ©2020 Europa Science Ltd.
While every care has been taken in the compilation of this magazine, errors or omissions are not the responsibility of the publishers or of the editorial staff. Opinions expressed are not necessarily those of the publishers or editorial staff. All rights reserved. Unless specifically stated, goods or services mentioned are not formally endorsed by Europa Science Ltd, which does not guarantee or endorse or accept any liability for any goods and/or services featured in this publication.
US copies: Scientific Computing World (ISSN 1356-7853/USPS No 018-753) is published bi-monthly for £180 per year by Europa Science Ltd, and distributed in the USA by DSW, 75 Aberdeen Rd, Emigsville PA 17318-0437. Periodicals postage paid at Emigsville PA. Postmaster: Send address corrections to Scientific Computing World PO Box 437, Emigsville, PA 17318-0437.
Cover: Matej Kastelic/
Shutterstock.com
Subscribe for free at
www.scientific-computing.com/subscribe
www.scientific-computing.com | @scwmagazine @scwmagazine February/March 2020 Scientific Computing World 3
two-part feature on data standards in the laboratory. The data standards discussion continues on page 17 with an article looking at ACD Labs’ strategy to maximise data value. On page 20, the modelling and simulation feature focuses on the use of optimisation strategies for engineers and designers. Increasingly sophisticated software makes it possible to combine optimisation with manufacturing processes such as additive manufacturing or to build complex multi-disciplinary optimisation pipelines.
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28
