Faster Population Counts Using AVX2 Instructions. (24th May 2017)

• Record Type: Journal Article
• Authors: Muła, Wojciech; Kurz, Nathan; Lemire, Daniel
• Is Part Of: Computer journal. Volume 61:Number 1(2018)
• Page Start: 111
• Page End: 120
• Publication Date: 2017-05-24
• Subjects: software performance -- SIMD instructions -- vectorization -- bitset -- Jaccard index; Computers -- Periodicals; 005.1
• DOI: 10.1093/comjnl/bxx046
• Languages: English
• ISSNs: 0010-4620

Abstract

Abstract: Counting the number of ones in a binary stream is a common operation in database, information-retrieval, cryptographic and machine-learning applications. Most processors have dedicated instructions to count the number of ones in a word (e.g. popcnt on ×64 processors). Maybe surprisingly, we show that a vectorized approach using SIMD instructions can be twice as fast as using the dedicated instructions on recent Intel processors. The benefits can be even greater for applications such as similarity measures (e.g. the Jaccard index) that require additional Boolean operations. Our approach has been adopted by LLVM: it is used by its popular C compiler (Clang).