March 2017

Big Bang Fair: Day Three

Author: Mirren White
Posted: 18 Mar 2017 | 22:36

Yesterday was Day Three of the Big Bang Fair, and it has kept delivering all the time. 

One of the most interesting things for me has been comparing this year to last year. For example, one of our most popular activities from last year, the supercomputing app, has been much quieter this year - but the beanbag sorting game has been a huge hit! 

Return to the Big Bang Fair

Author: Mario Antonioletti
Posted: 17 Mar 2017 | 07:54

The Big Bang Fair, billed as the UK's largest celebration of science, technology, engineering and maths for young people, provides our Supercomputing Sleuths event with an excellent opportunity to show the benefits of supercomputing to a host of young people through a number of targeted acitivities.

First day at BBF 2017

Author: Juan Rodriguez Herrera
Posted: 15 Mar 2017 | 22:04

During my period as a PhD student, I dealt with questions from academics, ranging from university students to full professors. However today at the Big Bang Fair (BBF) it was slightly different. This time the questions regarding what supercomputing is and what it involves came from pupils from different UK schools.

Back at the Big Bang Fair

Author: tpeet
Posted: 15 Mar 2017 | 09:57

ARCHER is back at the Big Bang Fair, the largest celebration of STEM (Science, Technology, Engineering and Maths) for young people in the UK. Held in Birmingham at the NEC over four days, there were over 70,000 visitors to the exhibition floor last year, so it will be a very busy time.

Spreading the love

Author: Adrian Jackson
Posted: 10 Mar 2017 | 13:54

Binding processes to coresThread and process binding

Note, this post was updated on the 23rd March 2017 to include how to bind threads correctly on Cray systems (aprun -cc rather than taskset)

Making sure threads and processes are correctly placed, or bound, on cores or processors is essential to ensure good performance for a range of parallel applications. 

This is not a new topic, and has been covered well by others before, ie http://www.glennklockwood.com/hpc-howtos/process-affinity.html. Generally this is just handled for you; if you're running an MPI program then your mpirun/mpiexec/aprun job launcher will do sensible process binding to cores.