2017 IEEE 24th Symposium on Computer Arithmetic (ARITH)最新文献

筛选
英文 中文
A Parallel Method for the Computation of Matrix Exponential Based on Truncated Neumann Series 基于截断诺伊曼级数的矩阵指数并行计算方法
2017 IEEE 24th Symposium on Computer Arithmetic (ARITH) Pub Date : 2017-07-24 DOI: 10.1109/ARITH.2017.23
V. Dimitrov, V. Ariyarathna, D. Coelho, L. Rakai, A. Madanayake, R. Cintra
{"title":"A Parallel Method for the Computation of Matrix Exponential Based on Truncated Neumann Series","authors":"V. Dimitrov, V. Ariyarathna, D. Coelho, L. Rakai, A. Madanayake, R. Cintra","doi":"10.1109/ARITH.2017.23","DOIUrl":"https://doi.org/10.1109/ARITH.2017.23","url":null,"abstract":"This paper introduces a new method for computing matrix exponential based on truncated Neumann series. The efficiency of the method is based on smart factorizations for evaluation of several Neumann series that can be done in parallel and divided across different processors with low communication overhead. A physical realization on FPGA is provided for proof-of-concept. The method is verified to be advantageous over the usual Horner's rule approach for polynomial evaluation. The hardware verification shows a reduction of 62% in time required for processing for series approximations with 9 terms. Software verification demonstrates a 30% reduction in time compared to Horner's rule and the trade-offs between using a higher precision approach is illustrated.","PeriodicalId":279178,"journal":{"name":"2017 IEEE 24th Symposium on Computer Arithmetic (ARITH)","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134069967","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 2
Efficient Leak Resistant Modular Exponentiation in RNS RNS中的高效防泄漏模幂
2017 IEEE 24th Symposium on Computer Arithmetic (ARITH) Pub Date : 2017-07-24 DOI: 10.1109/ARITH.2017.39
Andrea Lesavourey, C. Nègre, T. Plantard
{"title":"Efficient Leak Resistant Modular Exponentiation in RNS","authors":"Andrea Lesavourey, C. Nègre, T. Plantard","doi":"10.1109/ARITH.2017.39","DOIUrl":"https://doi.org/10.1109/ARITH.2017.39","url":null,"abstract":"The leak resistant arithmetic in RNS was introduced in 2004 to randomize RSA modular exponentiation. This randomization is meant to protect implementations on embedded device from side channel analysis. We propose in this paper a faster version of the approach of Bajard et al. in the case of right-to-left square-and-multiply exponentiation. We show that this saves roughly 30% of the computation when the randomization is done at each loop iteration. We also show that the level of randomization of the proposed approach is better than the one of Bajard et al. after a few number of loop iterations.","PeriodicalId":279178,"journal":{"name":"2017 IEEE 24th Symposium on Computer Arithmetic (ARITH)","volume":"85 2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127433717","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 2
Floating Point Tangent Implementation for FPGAs fpga的浮点切线实现
2017 IEEE 24th Symposium on Computer Arithmetic (ARITH) Pub Date : 2017-07-01 DOI: 10.1109/ARITH.2017.25
M. Langhammer, B. Pasca
{"title":"Floating Point Tangent Implementation for FPGAs","authors":"M. Langhammer, B. Pasca","doi":"10.1109/ARITH.2017.25","DOIUrl":"https://doi.org/10.1109/ARITH.2017.25","url":null,"abstract":"This paper presents an implementation of the floating-point (FP) tangent function, optimized for an FPGA containing hard floating point (HFP) DSP Blocks. This function inputs values in the interval [-π/2,π/2], uses the IEEE-754 single-precision (SP) format, and has an accuracy conforming to OpenCL requirements. The presented architecture is based on a combination of mathematical identities and properties of the tangent function in FP. The resultant design outperforms generic polynomial approximation methods targeting the same resource utilization spectrum, and provides better resource trade-offs than classical CORDIC-based implementations. The presented work is widely available as part of the Intel DSP Builder Advanced Blockset.","PeriodicalId":279178,"journal":{"name":"2017 IEEE 24th Symposium on Computer Arithmetic (ARITH)","volume":"76 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129914543","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
A New Multiplicative Inverse Architecture in Normal Basis Using Novel Concurrent Serial Squaring and Multiplication 一种新的基于并行串行平方和乘法的正基乘法逆结构
2017 IEEE 24th Symposium on Computer Arithmetic (ARITH) Pub Date : 2017-07-01 DOI: 10.1109/ARITH.2017.37
Amin Monfared, H. El-Razouk, A. Reyhani-Masoleh
{"title":"A New Multiplicative Inverse Architecture in Normal Basis Using Novel Concurrent Serial Squaring and Multiplication","authors":"Amin Monfared, H. El-Razouk, A. Reyhani-Masoleh","doi":"10.1109/ARITH.2017.37","DOIUrl":"https://doi.org/10.1109/ARITH.2017.37","url":null,"abstract":"Itoh and Tsujii proposed a fast algorithm for computing multiplicative inverses (inversions) over GF(2m) using normal bases by iterating single multiplications and cyclic shifts. Recently, the Itoh-Tsujii algorithm (ITA) has been modified to use two digit-level single multiplications. The improvements of the modified Itoh-Tsujii and its variant algorithms are based on reducing the computational latency at the expense of more area requirements. In this paper, we propose a new inversion architecture based on the classical IT algorithm (or improved one) utilizing a novel interleaved computations of two single multiplications and squarings at the digit-level. The new inverter outperforms previous modified Itoh-Tsujii algorithms (such as the Ternary Itoh-Tsujii and optimal 3-chain algorithms) in terms of its lower latency, higher throughput, and improved hardware efficiency. The efficiency of the proposed field inverter is demonstrated by comparisons based on application specific integrated circuits (ASIC) implementations results using the standard 65nm CMOS technology libraries.","PeriodicalId":279178,"journal":{"name":"2017 IEEE 24th Symposium on Computer Arithmetic (ARITH)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115491800","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 4
Flexible Fixed-Point Function Generation for FPGAs fpga灵活的定点函数生成
2017 IEEE 24th Symposium on Computer Arithmetic (ARITH) Pub Date : 2017-07-01 DOI: 10.1109/ARITH.2017.31
Matei Iştoan, B. Pasca
{"title":"Flexible Fixed-Point Function Generation for FPGAs","authors":"Matei Iştoan, B. Pasca","doi":"10.1109/ARITH.2017.31","DOIUrl":"https://doi.org/10.1109/ARITH.2017.31","url":null,"abstract":"Efficient fixed-point function implementation is critical in many FPGA application domains including convolutional neural networks, computer vision, and communication systems. In this work we focus on functions of the form xp, with p E{–1,–1/2,1/2} as part of a function generator targeting FPGAs. The generator implements architectures based on new but also existing algorithms. In this work we present three distinct methods implemented in this generator that outperform state-of-the-art implementations for certain configurations. Traditionally, fixed-point function implementation requires a normalization stage, compute and denormalization (reconstruction) of the result. The first proposed method implements the function holistically, thus saving the logic and latency required during the normalize and reconstruct stages. The second proposed method is based on a novel second order Taylor implementation. The third method is based on the cubic convergence of Halley's method, which is novel in this context. The proposed methods are compared and contrasted against state-of-the art implementations in the context of FPGA targets.","PeriodicalId":279178,"journal":{"name":"2017 IEEE 24th Symposium on Computer Arithmetic (ARITH)","volume":"230 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116387580","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 3
Resource Optimal Design of Large Multipliers for FPGAs fpga大型乘法器的资源优化设计
2017 IEEE 24th Symposium on Computer Arithmetic (ARITH) Pub Date : 2017-07-01 DOI: 10.1109/ARITH.2017.35
M. Kumm, Johannes Kappauf, Matei Iştoan, P. Zipf
{"title":"Resource Optimal Design of Large Multipliers for FPGAs","authors":"M. Kumm, Johannes Kappauf, Matei Iştoan, P. Zipf","doi":"10.1109/ARITH.2017.35","DOIUrl":"https://doi.org/10.1109/ARITH.2017.35","url":null,"abstract":"This work presents a resource optimal approach for the design of large multipliers for FPGAs. These are composed of smaller multipliers which can be DSP blocks or logic-based multipliers. A previously proposed multiplier tiling methodology is used to describe feasible solutions of the problem. The problem is then formulated as an integer linear programming (ILP) problem which can be solved by standard ILP solvers. It can be used to minimize the total implementation cost or to trade the LUT cost against the DSP cost. It is demonstrated that although the problem is NP-complete, optimal solutions can be found for most practical multiplier sizes up to 64x64. Synthesis experiments on relevant multiplier sizes show slice reductions of up to 47.5% compared to state-of-the-art heuristic approaches.","PeriodicalId":279178,"journal":{"name":"2017 IEEE 24th Symposium on Computer Arithmetic (ARITH)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128518548","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 12
ULPs and Relative Error ulp和相对误差
2017 IEEE 24th Symposium on Computer Arithmetic (ARITH) Pub Date : 2017-07-01 DOI: 10.1109/ARITH.2017.30
Marius Cornea
{"title":"ULPs and Relative Error","authors":"Marius Cornea","doi":"10.1109/ARITH.2017.30","DOIUrl":"https://doi.org/10.1109/ARITH.2017.30","url":null,"abstract":"The paper establishes several simple, but useful relationships between ulp (unit in the last place) errors and the corresponding relative errors. These can be used when converting between the two types of errors, ensuring that the least amount of information is lost in the process. The properties presented here were already useful in IEEE conformance proofs for iterative division and square root algorithms, and should be so again to numerical analysts both in 'hand' proofs of error bounds for floating-point computations, and in automated tools which carry out, or support deriving such proofs. In most cases, the properties shown herein establish tighter bounds than found in the literature. They also provide 'conversion' rules of finer granularity, at floating-point value level instead of binade level, and take into account the special conditions which occur at binade ends. For this reason, the paper includes a small, but non-negligible element of novelty.","PeriodicalId":279178,"journal":{"name":"2017 IEEE 24th Symposium on Computer Arithmetic (ARITH)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125128046","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
On Improving the Performance Per Area of ASTC with a Multi-output Decoder 用多输出解码器提高ASTC的单位面积性能
2017 IEEE 24th Symposium on Computer Arithmetic (ARITH) Pub Date : 2017-07-01 DOI: 10.1109/ARITH.2017.32
K. Rovers, Sam Elliott
{"title":"On Improving the Performance Per Area of ASTC with a Multi-output Decoder","authors":"K. Rovers, Sam Elliott","doi":"10.1109/ARITH.2017.32","DOIUrl":"https://doi.org/10.1109/ARITH.2017.32","url":null,"abstract":"ASTC is an efficient and flexible texture compression format but it is relatively costly to implement in hardware. By outputting multiple texels from a single encoded ASTC block, we will show an performance per area improvement of 25%.","PeriodicalId":279178,"journal":{"name":"2017 IEEE 24th Symposium on Computer Arithmetic (ARITH)","volume":"230 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117320934","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Fast Arithmetic Modulo 2^x p^y ± 1 快速算术模2^ xp ^y±1
2017 IEEE 24th Symposium on Computer Arithmetic (ARITH) Pub Date : 2017-07-01 DOI: 10.1109/ARITH.2017.15
Joppe W. Bos, Simon J. Friedberger
{"title":"Fast Arithmetic Modulo 2^x p^y ± 1","authors":"Joppe W. Bos, Simon J. Friedberger","doi":"10.1109/ARITH.2017.15","DOIUrl":"https://doi.org/10.1109/ARITH.2017.15","url":null,"abstract":"We give a systematic overview of techniques to compute arithmetic modulo 2x py ± 1 and propose improvements. This is useful for computations in the supersingular isogeny Diffie-Hellman (SIDH) key-exchange protocol which is one of the more recent contenders in the post-quantum public-key arena. One of the main computational bottlenecks in this cryptographic key-exchange protocol is computing modular arithmetic in a finite field defined by a prime of this special shape.Recent implementations already use this special prime shape to speed up the cryptographic implementations but it remains unclear if the choices made are optimal or if one can do better. Our overview shows that in the SIDH setting, where arithmetic over a quadratic extension field is required, the approaches based on Montgomery multiplication are to be preferred. Based on our results, we give selection criteria for such moduli and the outcome of our search reveals that there exist moduli which result in even faster implementations.","PeriodicalId":279178,"journal":{"name":"2017 IEEE 24th Symposium on Computer Arithmetic (ARITH)","volume":"77 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125001529","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 12
Large Scale Numerical Simulations of the Climate 气候的大尺度数值模拟
2017 IEEE 24th Symposium on Computer Arithmetic (ARITH) Pub Date : 2017-07-01 DOI: 10.1109/ARITH.2017.21
J. Rioual
{"title":"Large Scale Numerical Simulations of the Climate","authors":"J. Rioual","doi":"10.1109/ARITH.2017.21","DOIUrl":"https://doi.org/10.1109/ARITH.2017.21","url":null,"abstract":"The Met Office Hadley Centre for Climate Science and Services, based at the Met Office's Exeter HQ, provides world-class guidance on the science of climate change and is the primary focus in the UK for climate science. The Hadley Centre makes significant contributions to scientific literature and to a variety of climate science reports, including the International Panel on Climate Change (IPCC).In October 2014 the Government confirmed its investment of £97 million in a new high performance computing facility for the Met Office. The new Cray XC40 located at Exeter Science Park is the largest HPC system in the world dedicated to weather and climate research.I will first give an overview of how climate numerical experiments are organised worldwide through the coupled model intercomparison project (CMIP) under the auspices of the world climate research program (WCRP).The numerical results of these simulation campaigns are submitted to intense scrutiny by the scientific community and policymakers. Numerical reproducibility is therefore of paramount importance. I will explain the parameters of what numerical reproducibility means to our community and how we aim to achieve it.I will present the use of different types of floating point arithmetic in the models. Two examples are the use double-double precision for reproducible global sums and research on single precision algorithms for computational efficiency.Finally, I will look at some of the challenges to maintain numerical reproducibility in the exascale era.","PeriodicalId":279178,"journal":{"name":"2017 IEEE 24th Symposium on Computer Arithmetic (ARITH)","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132411431","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
相关产品
×
本文献相关产品
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信
小红书