Download presentation
Presentation is loading. Please wait.
Published byDavid Webb Modified over 9 years ago
1
COMMUNICATION COMMUNICATE COMMUNITY Henri Bal A PUBLIC-PRIVATE RESEARCH COMMUNITY
2
Agenda General introduction to COMMIT/ – (slides by Arnold Smeulders) Example collaboration NLeSC & COMMIT/: global climate modelling – NLeSC eSalsa project – COMMIT/ IV-e project
3
COMMIT/ The public-private research community for ICT- research of 60+ (non)profit & 20 science partners. Our mission is to deliver world-class ICT-research to create opportunities with more & more partners.
4
DIMENSIONS ICT-science use-inspired science Synergy between projects Dissemination tell the story outside science Valorization towards use in industry Internationalization NL is not alone in this field
5
SUCCESSES ICT-science Synergy SWEET Euro-Par 2014 Achievement award
6
SUCCESSES Dissemination Valorization Internationalization InfoSys and Figshare become partners, EU projects, EIT-ICT, …….
7
THE BIG FUTURE OF DATA We have presented 60 demonstrators: ICT Science Application Alternative application for four audiences:
8
The Big Future of Data Big Data are not always Big. Often they are: private interactive uncertain unordered new media In short they are overwhelming data.
9
The Big Future of Data Data-Science will surpass ICT-Science, that is our future.
10
IV-e: e-Infrastructure Virtualization for e-Science Applications (P20) e-Infrastructures become very complicated –Distributed & heterogeneous P20 tries to simplify the e-Infrastructure, ease programming and integration WP5 studies how to effectively use GPUs for data processing (imaging) and computing vrije Universiteit
11
Global Climate Modeling COMMIT/
12
GPU Computing Offload expensive kernels for Parallel Ocean Program (POP) from CPU to GPU –Many kernels, fairly easy to port to GPUs –Execution time becomes virtually 0 New bottleneck: I/O between CPU & GPU CPU host memory GPU device memory Host Device PCI Express link
13
Different methods for CPU-GPU communication Memory copies (explicit) – No overlap with GPU computation Device-mapped host memory (implicit) – Allows fine-grained overlap between computation and communication in either direction CUDA Streams or OpenCL command-queues – Allows overlap between computation and communication in different streams Any combination of the above
14
Problem ICT problem: – Which method will be most efficient for a given GPU kernel? Implementing all can be a large effort Solution: – Create a performance model that identifies the best implementation: Which strategy for overlapping computation and communication is best for the given program? Ben van Werkhoven, Jason Maassen, Frank Seinstra & Henri Bal: Performance models for CPU-GPU data transfers, CCGrid2014 (nominated for best-paper-award, top 1%)
15
MOVIE
16
Example result MeasuredModel
17
Discussion Good example of use-inspired Computer Science research Many other application domains that use accelerators face the same problem – Digital forensics, water management, astronomy … All details: 27 October 2014 1 October 2014
19
Different GPUs (state kernel)
20
Different GPUs (buoydiff)
21
Global Climate Modeling Understand future local sea level changes Needs high-resolution simulations Combine two approaches: – Distributed computing (multiple resources) Ibis couples models for land, ice, ocean, atmosphere – GPUs(Graphics processing units) COMMIT/
22
Comes with spreadsheet
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.