| Dresden 20 June 2008
The performance was not exactly a "Battle-Rap", but it came very close. Wolfgang Gentzsch and Dieter Kranzmueller gave at the closing day in Dresden "battle" about Grid computing versus Supercomputing. Everybody knows Wolfgang Gentzsch as mr Grid: Genias, Gridware, SUN Grid engine, D-Grid. But he changed camps and is now dissemination manager of DEISA the European supercomputer collaboration project. Dieter Kranzmueller was active in supercomputing for some time, but today is involved in the European Grid Initiative. Wolfgang kicked of with "Grids are dead" but Dieter replied immediately with recent Grid successes, such as calculations for the Avian Flu, that were set-up very fast and did lead to useful results, and WISDOM, that helps to contribute to solving Malaria.
|
|
|
Wolfgang seemed to hesitate for a moment, but then really started a flood of anti-Grid arguments: no progress in ten years, confusion about what a Grid is, lack of standards; and Grids are under attacks by Clouds. That did not impress Dieter much. No progress in 10 years? Supercomputer applications still use 30 year old Fortran? Standards: perhaps Wolfgang did not stay log enough at an Open grid forum meeting to note a dozen or so standards have come out during the past few years. And yes the top speed of supercomputers is increasing, but applications are less and less efficient. Supercomputers just do not scale very well: they perform Ok on the artificial Linpack benchmark, but that is it.
Of course, they agreed in the end that there was room for both grids and supercomputers: there should be one integrated European HPC infrastructure, with large supercomputers at the top. The supercomputers should be part of a supercomputing grid to allow users to choose the right supercomputer for their applications, for more loosely coupled applications, such as CERN's LHC, a standard grid of clusters should be available.
|
