Astron is developing the LOFAR radio telescope (http://www.lofar.nl). This telescope will exists of approximately 5.000 small dipole antennas, organized in 36 antenna groups spread over and area with a 80km radius. The data generated by these antennas (4 - 20Gb/s per group) will be processed by a Blue Gene supercomputer located at Groningen University. Data transfer from the antenna groups to the supercomputer will take place with the Ethernet protocol over a (dedicated) fiber network.

This network will be largely realized with commercial off the shelf products. These products will be acquired via a public tender. My role is the design of the reference network and selecting the type of product needed. This information will be used to create the public tender, which will also be partly written by me.

Additionally, several additional antenna groups will be located abroad. These stations will be connected via fast Internet connections or dedicated fiber lines “light paths”. Since all data is processed online, the delay caused by the geological distance is creating some interesting challenges.

2 other large programs involving the same skills are also running at Astron. For the Expres project (http://www.expres-eu.org) I'm involved in Joint Activity Research task 1. In this task we research how a future VLBI infrastructure will look like. Involved technologies are “light paths”, long haul 10Gb/s networks and grid computing.

The other large program is the Square Kilometer Array design study (http://www.skatelescope.org). This is a future telescope employing a detecting area of a square kilometer. LOFAR is working around 100 Mhz band. SKA will be operating in the Ghz band. This means that the amount of data generated by SKA is at least 10 times higher than for LOFAR. At the moment it is not feasible with existing infrastructure components. But we expect it will be in the next couple of years.