作者: WenMei Ong , Vishnu Monn Baskaran , Poh Kit Chong , K. K Ettikan , Keh Kok Yong
DOI: 10.1109/ICOS.2013.6735037
关键词: Bloom filter 、 Set (abstract data type) 、 CUDA 、 Parallel computing 、 Speedup 、 Search algorithm 、 String searching algorithm 、 Computer science 、 Parallel algorithm 、 String (computer science)
摘要: This paper analyzes the underlying architecture of a serial Bloom filter string searching algorithm to identify performance impact this for large datasets. Then, parallel multi-core driven using software application threads is studied as benchmark. Experimental results suggest that set 10 million strings, exhibits speedups up 3.3× against algorithm, when an 8-logical processor architecture. To further improve speedup, many-core proposed Compute Unified Device Architecture (CUDA) computing platform. The segments list into blocks words and in generating bit table process, which maximizes computational sustains consistent results. extends speedup 5.5× 256-core CUDA graphics processing unit