Memory Memory Memory Compute Processor Compute Processor Compute Processor Interconnection

Many scienti c applications that run on today s multiprocessors such as weather forecast ing and seismic analysis are bottlenecked by their le I O needs Even if the multiprocessor is con gured with su cient I O hardware the le system software often fails to provide the available bandwidth to the application Although libraries and enhanced le system interfaces can make a signi cant improvement we believe that fundamental changes are needed in the le server software We propose a new technique disk directed I O to allow the disk servers to determine the ow of data for maximum performance Our simulations show that tremendous performance gains are possible Indeed disk directed I O provided consistent high performance that was largely independent of data distribution obtained up to of peak disk bandwidth and was as much as times faster than traditional parallel le systems