Near Memory Similarity Search on Automata Processors
University of Washington
arXiv:1608.03175 [cs.DC], (9 Aug 2016)
@article{lee2016near,
title={Near Memory Similarity Search on Automata Processors},
author={Lee, Vincent T. and Kotalik, Justin and Mundo, Carlo C. Del and Alaghi, Armin and Ceze, Luis and Oskin, Mark},
year={2016},
month={aug},
archivePrefix={"arXiv"},
primaryClass={cs.DC}
}
Embedded devices and multimedia applications today generate unprecedented volumes of data which must be indexed and made searchable. As a result, similarity search has become a critical idiom for many modern data intensive applications in natural language processing (NLP), vision, and robotics. At its core, similarity search is implemented using k-nearest neighbors (kNN) where computation consists of highly parallel distance calculations and a global top-k sort. In contemporary von-Neumann architectures, kNN is bottlenecked by data movement limiting throughput and latency.
August 16, 2016 by hgpu
