Acceleration of Nonnumeric Operations Using Hardware Support for the Ordered Table Hashing Algorithms
Issue No. 09 - September (2002 vol. 51)
<p><b>Abstract</b>—This paper introduces a new approach to acceleration of nonnumeric, database, and information retrieval operations. While traditional techniques accelerate the most time-critical high-level software constructs, we propose novel low-level primitives and demonstrate how these primitives improve database operations. Radix sorting, hashing, and bit-vector operations are used to develop a new class of nonnumeric algorithms—<b>OTHER</b> (<b>O</b>rdered <b>T</b>able <b>H</b>ashing and <b>R</b>adix sort algorithms)—based on low-level hashing operations Init, Mark, and Scan. We have proposed and evaluated two hardware accelerators for OTHER algorithms. It is shown that a low complexity hardware support (less than 10K transistors) can significantly improve the performance of nonnumeric operations.</p>
Database operations, hashing, sorting, searching, nonnumeric processing, hardware accelerators.
Emil Jovanov, Veljko Milutinovic, Ali R. Hurson, "Acceleration of Nonnumeric Operations Using Hardware Support for the Ordered Table Hashing Algorithms", IEEE Transactions on Computers, vol. 51, no. , pp. 1026-1040, September 2002, doi:10.1109/TC.2002.1032623