A hybrid CPU-FPGA system for high throughput (10Gb/s) streaming document classification

Vanderbauwhede, W. , Frolov, A., Chalamalasetti, S. R. and Margala, M. (2014) A hybrid CPU-FPGA system for high throughput (10Gb/s) streaming document classification. Computer Architecture News, 41(5), pp. 53-58. (doi: 10.1145/2641361.2641370)

Full text not currently available from Enlighten.

Publisher's URL: http://dx.doi.org/10.1145/2641361.2641370

Abstract

Processing large volumes of information in real time requires large amounts of computational power, which consumes a significant amount of energy. With the rise in the amount of data produced, energy-efficient high-performance information processing systems are becoming a necessity. We present a hybrid CPU-FPGA system for high-throughput classification of streams of textual documents (e.g. emails or web pages). The current system parses the document stream using a multicore CPU and performs classification on the parsed stream using Field-Programmable Gate Arrays (FPGAs). As an example, we demonstrate a Naive Bayes classifier on the TREC Aquaint dataset. Our current solution can classify 10Gb/s internet traffic in real time. Our aim is to increase the throughput to 100Gb/s by incorporating the parser into the FPGA design.

Item Type:Articles
Status:Published
Refereed:Yes
Glasgow Author(s) Enlighten ID:Vanderbauwhede, Professor Wim and Frolov, Mr Anton
Authors: Vanderbauwhede, W., Frolov, A., Chalamalasetti, S. R., and Margala, M.
College/School:College of Science and Engineering > School of Computing Science
Journal Name:Computer Architecture News
Publisher:ACM
ISSN:0163-5964
ISSN (Online):1943-5851

University Staff: Request a correction | Enlighten Editors: Update this record