The problem

Research in some areas (combinatorial problems, computational biology, computer simulation, etc.) involves intensive use of computational resources. Traditionally researchers in such areas have used supercomputers in supercomputer centers in a time and node sharing fashion. There are some problems with this approach:

• Supercomputers tend to be very expensive to acquire (millions of dollars) and to maintain. This fact is due partly to the fact that the biggest client for companies building supercomputers is/was the Government which was generously encouraging research in this domain. Also the competition used to be quite weak.
• Because supercomputers are expensive, parts required to repair them are expensive also, thus the maintenance costs are high.
• A big number of researchers have access to one supercomputer competing for CPU time (and occasionally other resources), thus quite often one has to wait weeks until his/her tasks are run.
• Is expensive to run something on a supercomputer (since supercomputers are themselves expensive), thus quite big grants are required to do anything serious.