Why do we need PGI http:www.unicore.eu

1
Why do we need PGI?http//www.unicore.eu

• Shahbaz Memonm.memon_at_fz-juelich.deJülich
  Supercomputing Centre (JSC)

2
Guiding Principles

• Open source under BSD license with software
  hosted on SourceForge
• Standards-based OGSA-conform, WS-RF 1.2
  compliant
• Open, extensible Service-Oriented Architecture
  (SOA)
• Interoperable with other Grid technologies
• Seamless, secure and intuitive following a
  vertical end-to-end approach
• Mature Security X.509, optional proxy and VO
  support
• Workflow support tightly integrated while being
  extensible for different workflow languages and
  engines for domain-specific usage
• Application integration mechanisms on the client,
  services and resource level
• Variety of clients graphical, command-line, API,
  portal, etc.
• Quick and simple installation and configuration
• Support for many operating systems (Windows,
  MacOS, Linux, UNIX) and batch systems
  (LoadLeveler, Torque, SLURM, LSF, OpenCCS)
• Implemented in Java to achieve platform-independen
  ce

3
Architecture PGI Adoption
scientific clientsand applications
authentication
emerging standardinterfaces
Grid services hosting
GLUE 2.0
job incarnation authorization
parallel scientific jobs of multiple end-users
on target systems
4

PGI Perspective

• Answers to the question Why do we need PGI?
• UNICORE is primarily driven by High Performance
  Computing (HPC) needs in production
  infrastructures
• e.g. HPC-driven EU DEISA, Russian-Belarus
  SKIF-Grid,
• Compliance to PGI enables the use of HPC
  applications in conjunction with rather High
  Throughput Computing (HTC) driven
  infrastructures (i.e. EGEE, NorduGrid, NGS,)
• Avoiding workarounds/hacks to achieve
  interoperability with HTC-driven middleware on a
  pair-wise fashion (i.e. gLite/EGEE,
  ARC/NorduGrid) in order to run HTC/HPC scientific
  workflows
• Enable scientists to use HTC infrastructures/resou
  rces for small-scale evaluation jobs before doing
  full-blown production with up to 16.000 cores and
  more on large-scale HPC resources
• Enable HTC infrastructure users with an easy way
  to use HPC