skip to navigation skip to content
Thu 1 Jun 2017
09:00 - 17:00

Venue: University Information Services, Phoenix Teaching Room 1, New Museums Site

Provided by: University Information Services - Digital Literacy Skills


Booking

Bookings cannot be made on this event (Event is completed).


Other dates:

No more events

[ Show past events ]



Register interest
Register your interest - if you would be interested in additional dates being scheduled.


Booking / availability

High Performance Computing: Programming GPU using Open ACC
Prerequisites

Thu 1 Jun 2017

Description

This course aims to provide a basic knowledge of GPU programming using OpenACC directives. The course is very hands-on oriented, aiming to give to you the opportunity to practice and experiment from the very beginning.

Target audience
  • All current Cambridge University members (departments and colleges)
  • Especially University alumni (PhD) and researchers who already have access to HPCS computing facilities
  • Further details regarding eligibility criteria are available
Prerequisites
Sessions

Number of sessions: 2

# Date Time Venue Trainer
1 Thu 1 Jun 2017   09:00 - 12:30 09:00 - 12:30 University Information Services, Phoenix Teaching Room 1, New Museums Site map Filippo Spiga
2 Thu 1 Jun 2017   13:30 - 17:00 13:30 - 17:00 University Information Services, Phoenix Teaching Room 1, New Museums Site map Filippo Spiga
Topics covered
  • General introduction about accelerated computing
  • APOD: Assess, Parallelize, Optimize, Deploy
  • OpenACC basic syntax:
    • parallel loop directive
    • define data regions
    • data clause
    • kernel construct
    • reduction clause
  • Managing parallelism: gang, worker, vector
  • Understand compiler output
  • Mapping OpenACC to CUDA
    • Optimization techniques:
    • kernel fusion
    • async & wait clauses
    • pipelining
    • advanced data layouts
Format
  • Presentations and practicals
Taught using
  • OpenACC (PGI compiler installed on HPCS systems) and PuTTY on MCS Windows
Notes
  • HPCS users will have the opportunity to perform experiments directly to Wilkes
  • Participants are encouraged to bring their own scientific applications and discuss with the trainers about parallelization strategies and opportunities
Duration
  • One full day
Related courses
Themes

Booking / availability