Showing posts with label risk. Show all posts
Showing posts with label risk. Show all posts

Thursday 26 May 2011

Poll: legacy software and future HPC applications

I've added a new quick survey to the HPC Notes blog: "Which do you agree with more for developing the next generation of HPC applications?"

Is the argument about "protecting our invetsment" a load of nonsense? Or is throwing it all away and starting again irresponsible?

I've only allowed the two extremes as voting options - I know you might want to say "both" - but choose one!

See top right of the blog home page. Vote away ...

For clues on my own views, and some good audience debate, see the software panel video recording from the recent NCSA PSP annual meeting here:

Blog on the topic to follow shortly ... after some of your votes have been posted :-)

Thursday 28 January 2010

Are we taking supercomputing code seriously?

[Article by me on ZDNet UK, 28 January, 2010]

The supercomputing programs behind so much science and research are written by people who are not software pros ...

Thursday 18 June 2009

When supercomputing benchmarks fail to add up

[Article by me on ZDNet UK, 18 June, 2009]

Using benchmarks to choose a supercomputer is more complex than just picking the fastest system ...

Friday 20 March 2009

Are supercomputers just better liars?

[Article by me on ZDNet UK, 20 March, 2009]

Supercomputers may be far more powerful than ordinary machines, but that does not make their predictions infallible ...

Thursday 5 February 2009

What to do if your supercomputing supplier fails

[Article by me on ZDNet UK, 5 February, 2009]

High-performance computing providers often live on the edge — technologically and financially. But if your supplier fails, it need not be a disaster ...