Introduction to ARCHER and Cray MPI

An introduction to using the new national supercomputer ARCHER and system-specific features of the Cray MPI library. Familiarity with parallel programming is assumed. The course is open to all, but is mainly targeted at existing HECToR users.

Details

The new ARCHER supercomputer, managed by EPSRC and operated by EPCC at the University of Edinbugh, is a 72,192 core Cray XC30 system. Although the overall software environment should be familiar to existing users of the HECToR Cray XE6 system, there are several important differences between ARCHER and HECToR. For example, ARCHER uses 12-core 2.7GHz Intel E5-2697 v2 (Ivy Bridge) series processors as opposed to HECToR's 16-core 2.3GHz AMD Opteron (Interlagos) processors. The network is the new Cray Aries interconnect, and ARCHER now supports the Intel compiler suite.

This course will give an overview of the hardware and software environment on ARCHER, focusing on areas where it differs from HECToR. ARCHER-specific features of the MPI library will be explained, for example ways to make the most of the capabilities of the Aries interconnect. The final session will focus on helping users to port applications to ARCHER. Hands-on practical exercises will form an integral part of the course. Example codes will be supplied but we hope that many users will work on their own applications in the practcial sessions.

This course is free to all academics.

Intended learning outcomes

On completion of this course students should be able to:

  • Understand the ARCHER hardware environment.
  • Compile and run parallel programs on ARCHER.
  • Exploit ARCHER-specific features of the MPI library in their own codes.
  • Port applications from HECToR to ARCHER.

Pre-requisites

Attendees are expected to have prior experience of parallel programming, for example on HECToR.

Timetable

Details are subject to change, but start, end and break times will stay the same.

Day 1:

  • 09:30 LECTURE: ARCHER hardware
  • 10:30 PRACTICAL: Logging on, compiling and submitting jobs
  • 11:00 BREAK: Coffee
  • 11:30 LECTURE: Compiling for the ARCHER hardware
  • 12:15 PRACTICAL: Efficient compilation
  • 12:45 BREAK: Lunch
  • 14:00 LECTURE: ARCHER processors
  • 14:45 PRACTICAL: Using the Intel Ivy Bridge CPU
  • 15:30 BREAK: Tea
  • 16:00 LECTURE: Software environment
  • 16:45 PRACTICAL: ARCHER tools
  • 17:30 CLOSE

Day 2

  • 09:30 LECTURE: ARCHER MPI library
  • 10:30 PRACTICAL: Ping-pong exericse
  • 11:00 BREAK: Coffee
  • 11:30 LECTURE: Efficient use of MPI on ARCHER
  • 12:15 PRACTICAL: MPI optimisations
  • 12:45 BREAK: Lunch
  • 14:00 PRACTICAL: Porting your own applications
  • 15:30 BREAK: Tea
  • 16:00 PRACTICAL: Porting your own applications (contd)
  • 17:30 CLOSE

Registration

Please use the registration page to register for ARCHER courses.

Questions?

If you have any questions please contact the ARCHER Helpdesk.