PART: A Run-Time Library for Tape-Resident Data

Gokhan Memik, Mahmut Taylan Kandemir, Alok Choudhary, and Valerie E. Taylor

Abstract

Over the last decade, processors have made enormous gains in speed. But increase in the speed of the secondary and tertiary storage devices could not cope with these gains. The result is that the secondary and tertiary storage access times dominate execution time of data intensive computations. Therefore, in scientific computations, efficient data access functionality for data stored in secondary and tertiary storage is a must. In this extended abstract, we give an overview o f PART, a runtime library that can be used in applications that process tape-res ident data. We present user interface and underlying optimization strategy. We a lso discuss performance improvements provided by the library on the High Perform ance Storage System (HPSS). The preliminary results reveal that the optimization s can improve response times by up to 97:2%.

Gzipped Postscript version of the paper