hpcx power for the grid dr alan d simpson hpcx project director epcc technical director
TRANSCRIPT
HPCxPower for the Grid
Dr Alan D SimpsonHPCx Project Director
EPCC Technical Director
2September 2003HPCx
HPCx Overview
• UK’s major HPC facility, primarily funded by EPSRC• £53M/6 year contact awarded to UoE HPCX Ltd
– wholly-owned subsidiary of University of Edinburgh– work subcontracted to CCLRC (DL), EPCC and IBM
• Largest academic supercomputer in Europe– doubling in performance every 2 years
3September 2003HPCx
HPCx Objectives
• Deliver capability computing for world-leading science
• Capability Computing– jobs which use a significant fraction of the
resource, eg, at least 512 CPUs
• Collaboration between HPCx and users through the Terascaling process
• Maximise benefits to the UK’s computational science and engineering community
• Forging links between HPC and e-Science• High quality support is the key to success
4September 2003HPCx
Partnership
• EPCC and CCLRC– are partners in C3ES (Consortium for Capability
Computing and e-Science)– underpinned by MoU between UoE and CCLRC– combines Europe’s foremost academic HPC,
e-Science and technology transfer centres– virtual organisation facilitated by Access Grid– significant experience of:
• operating national HPC services• developing capability applications
– the strongest UK partnership ever to support scientific computing
5September 2003HPCx
Virtual Organisation
OutreachLife sciencesNew applications
Applications SupportHelpdesk
Training Liaising with users
Users
Technology
Software EngineeringUnderpinning technology Grid/e-
ScienceSystems & NetworkingFlexible and responsive capability computing
service Smooth transitions between phases
Terascaling Capability applicationsScalable algorithms Performance
optimisation
• Dual-centre functional support teams
6September 2003HPCx
HPCx Utilisation
• successful first 9 months• >75% utilisation for last 6 months• capability usage has increased to 35%
0
100000
200000
300000
400000
500000
600000
700000
800000
Jan-03 Feb-03 Mar-03 Apr-03 May-03 Jun-03 Jul-03 Aug-03
Usage
>1024 CPUs
1024 CPUs
512 CPUs
256 CPUs
128 CPUs
64 CPUs
32 CPUs
16 CPUs
8 CPUs
7September 2003HPCx
HPCx and the Grid
• Key responsibility for Software Engineering team– led by Dr Stephen Booth
• who is also responsible for EPCC’s Grid operations
• HPCx is committed to support access via Grid– currently provided through Globus 2– Globus 3 support when appropriate
• HPCx is key part of UK collaboration with Extensible Teragrid Facility project in the US– promoting UK science on the world stage
8September 2003HPCx
ETF Collaboration
• Focus is exploiting unique features of Grid + HPC systems for capability computing– `HPCy-class’ applications
• Initial experiment planned for SC2003– RealityGrid computational steering– HPCx is major compute resource
• Current challenges are:– network bandwidth– lack of direct network connections to compute
nodes• developing port-forwarding software to allow Globus IO
connections to batch jobs
9September 2003HPCx
Summary
• HPCx builds on significant complementary experience at EPCC and DL
• Very successful start– …with capability usage already up to 35%
• Committed to e-Science and the Grid– strong links with NeSC and CCLRC e-Science Centre– ETF experiment at SC2003
• HPCx is focussed on capability computing– world-class service for world-class research