FEATURES & COMMENTARY
To the editor:
Congratulations for placing an article about GSN in HPCWire. This will certainly work out as being very positive for the future of GSN.
However I have to make some correction on the point you make about HIPPI in Europe.
In 1989 CERN started already to do some work on HIPPI and in 1994 we had a small network with a NSC switch. Due to this advancement we were invited in 1994 to build a demonstration network at HPCN in Munich. This demonstration network with 5 switches and 14 nodes and very attractive demonstrations made HIPPI known in Europe. In 1995 CERN repeated this exercise at HPCN in Europe, using this occassion to build the EHUG HIPPI (User Group) today known as HNF-Europe.
At CERN we made the same year an 8 Km link between a physics experiment and the computer centre, at the same time debugging Serial-HIPPI. It was the start of a network that had at some time over 50 nodes.
As far as I know Stuttgard University started with HIPPI after both of our demonstration networks at HPCN. For the long distance coupling between the two supercomputer centers they still used ATM for a long time.
As I have been strongly involved myself in the different projects around HIPPI and GSN at CERN I can confirm that the dates mentioned above are correct. And as such I would highly appreciate that the credit for bringing HIPPI and GSN to Europe will be corrected and gets on the name of CERN.
Arie Van Praag
The viewpoints in this article are those of its author and not necessarily those of the publisher or staff of HPCwire.