dc.contributor.author | Umuroglu, Yaman | |
dc.contributor.author | Rasnayake, Lahiru | |
dc.contributor.author | Själander, Magnus | |
dc.date.accessioned | 2019-02-15T09:59:07Z | |
dc.date.available | 2019-02-15T09:59:07Z | |
dc.date.created | 2018-06-24T14:35:40Z | |
dc.date.issued | 2018 | |
dc.identifier.isbn | 000-0-0000-0000-0 | |
dc.identifier.uri | http://hdl.handle.net/11250/2585656 | |
dc.description.abstract | Matrix-matrix multiplication is a key computational kernel for numerous applications in science and engineering, with ample parallelism and data locality that lends itself well to high-performance implementations. Many matrix multiplication- dependent applications can use reduced-precision integer or fixed- point representations to increase their performance and energy efficiency while still offering adequate quality of results. However, precision requirements may vary between different application phases or depend on input data, rendering constant-precision solutions ineffective. We present BISMO, a vectorized bit- serial matrix multiplication overlay for reconfigurable computing. BISMO utilizes the excellent binary-operation performance of FPGAs to offer a matrix multiplication performance that scales with required precision and parallelism. We characterize the resource usage and performance of BISMO across a range of parameters to build a hardware cost model, and demonstrate a peak performance of 6.5 TOPS on the Xilinx PYNQ-Z1 board. BISMO is open-sourced on GitHub. | nb_NO |
dc.language.iso | eng | nb_NO |
dc.publisher | Institute of Electrical and Electronics Engineers (IEEE) | nb_NO |
dc.relation.ispartof | The International Conference on Field-Programmable Logic and Applications (FPL) | |
dc.title | BISMO: A Scalable Bit-Serial Matrix Multiplication Overlay for Reconfigurable Computing | nb_NO |
dc.title.alternative | BISMO: A Scalable Bit-Serial Matrix Multiplication Overlay for Reconfigurable Computing | nb_NO |
dc.type | Chapter | nb_NO |
dc.description.version | submittedVersion | nb_NO |
dc.identifier.doi | 10.1109/FPL.2018.00059 | |
dc.identifier.cristin | 1593478 | |
dc.description.localcode | © 2018 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. | nb_NO |
cristin.unitcode | 194,63,10,0 | |
cristin.unitname | Institutt for datateknologi og informatikk | |
cristin.ispublished | true | |
cristin.fulltext | preprint | |
cristin.qualitycode | 1 | |