检索结果-内蒙古大学图书馆

Structure optimization method based on automatic vectorization

EVOLUTIONARY INTELLIGENCE 2020年第1期13卷 51-58页

作者： Li, Yu-ping Guo, Zhan-jie Liu, Hui Shangqiu Normal Univ Sch Informat Technol Shangqiu 476000 Peoples R China Zhengzhou Tech Coll Dept Elect & Elect Engn Zhengzhou 450121 Peoples R China Informat Engn Univ Collge Informat Syst Engn Zhengzhou 450001 Peoples R China

Structure is used more extensively used in program such as scientific computing. But the non-continuity and the non-aligment of vectorization structure array have a dramatic influence on the efficiency of program's vectorizaton. To reduce the access to these addresses during the simd vectorization, a structure peeling model is proposed based on the structure which combines domain access affinity and domain data type. At the same time, to meet t the requirement of memory access continuity and alignment in the vectorization of structured array, an address conversion method is proposed which structure arrays are mapped one by one map to two-dimensional arrays, further reducing the failure rate of cache. By using the test suites of gcc_vec, spec2000 and spec2006, the experimental results on the compiler of automatic vector show that the performance of optimized method can be improved by more than 8%.

关键词： Structure peeling Address mapping simd vectorization Access affinity

来源：评论

学校读者我要写书评

暂无评论

HACLxN: Verified Generic simd Crypto (for all your favorite platforms) 20

HACLxN: Verified Generic SIMD Crypto (for all your favorite ...

引用

ACM SIGSAC Conference on Computer and Communications Security (ACM CCS)

作者： Polubelova, Marina Bhargavan, Karthikeyan Protzenko, Jonathan Beurdouche, Benjamin Fromherz, Aymeric Kulatova, Natalia Zanella-Beguelin, Santiago Inria Paris Paris France Microsoft Res Redmond WA USA Mozilla San Francisco CA USA Carnegie Mellon Univ Pittsburgh PA 15213 USA

ISBN: (纸本)9781450370899

We present a new methodology for building formally verified cryptographic libraries that are optimized for multiple architectures. In particular, we show how to write and verify generic crypto code in the F* programming language that exploits single-instruction multiple data (simd) parallelism. We show how this code can be compiled to platforms that support vector instructions, including ARM Neon and Intel AVX, AVX2, and AVX512. We apply our methodology to obtain verified vectorized implementations on all these platforms for the ChaCha20 encryption algorithm, the Poly1305 one-time MAC, and the SHA-2 and Blake2 families of hash algorithms. A distinctive feature of our approach is that we aggressively share code and verification effort between scalar and vectorized code, between vectorized code for different platforms, and between implementations of different cryptographic primitives. By doing so, we significantly reduce the manual effort needed to add new implementations to our verified library. In this paper, we describe our methodology and verification results, evaluate the performance of our code, and describe its integration into the HACL* crypto library. Our vectorized code has already been incorporated into several software projects, including the Firefox web browser.

关键词： formal verification simd vectorization cryptographic library

来源：评论

学校读者我要写书评

暂无评论

CSR2: A New Format for simd-accelerated SpMV 20

CSR2: A New Format for SIMD-accelerated SpMV

引用

20th IEEE/ACM International Symposium on Cluster, Cloud and Internet Computing (CCGrid)

作者： Bian, Haodong Huang, Jianqiang Dong, Runting Liu, Lingbin Wang, Xiaoying Qinghai Univ Dept Comp Technol & Applicat Xining Peoples R China Tsinghua Univ Dept Comp Sci & Technol Beijing Peoples R China

ISBN: (纸本)9781728160955

SpMV (Sparse matrix-vector multiplication) has attracted the attention of researchers in related fields at home and abroad. Of course, improving SpMV performance has also been a research hot spot for researchers in related fields. In this paper, we propose a new sparse matrix storage format CSR2 (Compressed Sparse Row 2) suitable for simd (Single Instruction Multiple Data)-accelerated SpMV. First, the format operation of CSR2 is easy to implement and has a low overhead of conversion. Second, CSR2 is a new single format and suitable for use on processor platforms with simd vectorization. We compare the SpMV algorithm based on CSR21 with the one based on the current most advanced single format CSR5 (Compressed Sparse Row 5) on two mainstream high-performance processors: Intel Core i7-7700HQ CPU and Intel Xeon CPU E5-2670 v3. We choose 10 sets of regular matrices and 3 sets of irregular matrices to be used as benchmark suit. Experiments show that for the 13 sets of regular and irregular matrices in the benchmark suite, CSR2 has an average performance improvement of more than 50% compared to CSR5 (up to 125% on Intel Core i7-7700HQ CPU and 303% on Intel Xeon CPU E5-2670 v3). For applications with multiple iterations, in reality, using our CSR2 can bring low-overhead format conversion and high-throughput computing performance.

关键词： SpMV simd vectorization CSR2 CSR5 Storage Formats CPU

来源：评论

学校读者我要写书评

暂无评论

An optimized FM-index library for nucleotide and amino acid search

引用

ALGORITHMS FOR MOLECULAR BIOLOGY 2021年第1期16卷 25-25页

作者： Anderson, Tim Wheeler, Travis J. Univ Montana Dept Comp Sci Missoula MT 59812 USA

Background: Pattern matching is a key step in a variety of biological sequence analysis pipelines. The FM-index is a compressed data structure for pattern matching, with search run time that is independent of the length of the database text. Implementation of the FM-index is reasonably complicated, so that increased adoption will be aided by the availability of a fast and flexible FM-index library. Results: We present AvxWindowedFMindex (AWFM-index), a lightweight, open-source, thread-parallel FM-index library written in C that is optimized for indexing nucleotide and amino acid sequences. AWFM-index introduces a new approach to storing FM-index data in a strided bit-vector format that enables extremely efficient computation of the FM-index occurrence function via AVX2 bitwise instructions, and combines this with optional on-disk storage of the index's suffix array and a cache-efficient lookup table for partial k-mer searches. The AWFM-index performs exact match count and locate queries faster than SeqAn3's FM-index implementation across a range of comparable memory footprints. When optimized for speed, AWFM-index is similar to 2-4x faster than SeqAn3 for nucleotide search, and similar to 2-6x faster for amino acid search;it is also similar to 4x faster with similar memory footprint when storing the suffix array in on-disk SSD storage. Conclusions: AWFM-index is easy to incorporate into bioinformatics software, offers run-time performance parameterization, and provides clients with FM-index functionality at both a high-level (count or locate all instances of a query string) and low-level (step-wise control of the FM-index backward-search process). The open-source library is available for download at https://githubscom/TravisWheelerLab/AvxWindowFmlndex.

关键词： FM-index String matching simd vectorization

来源：评论

学校读者我要写书评

暂无评论

TweTriS: Twenty trillion-atom simulation

引用

INTERNATIONAL JOURNAL OF HIGH PERFORMANCE COMPUTING APPLICATIONS 2019年第5期33卷 838-854页

作者： Tchipev, Nikola Seckler, Steffen Heinen, Matthias Vrabec, Jadran Gratl, Fabio Horsch, Martin Bernreuther, Martin Glass, Colin W. Niethammer, Christoph Hammer, Nicolay Krischok, Bernd Resch, Michael Kranzlmueller, Dieter Hasse, Hans Bungartz, Hans-Joachim Neumann, Philipp Tech Univ Munich Dept Informat Garching Germany Tech Univ Berlin Thermodynam & Proc Engn Berlin Germany AUIS Engn Dept Sulaimani Iraq STFC Daresbury Lab Sci Comp Dept Warrington Cheshire England High Performance Comp Ctr Stuttgart HLRS Stuttgart Germany Helmut Schmidt Univ Mech Engn Hamburg Germany Leibniz Supercomp Ctr Garching Germany Univ Stuttgart Inst High Performance Comp Stuttgart Germany TU Kaiserslautern Lab Engn Thermodynam LTD Kaiserslautern Germany Univ Hamburg Dept Informat Bundesstr 45a D-20146 Hamburg Germany

Significant improvements are presented for the molecular dynamics code ls1 mardyn - a linked cell-based code for simulating a large number of small, rigid molecules with application areas in chemical engineering. The changes consist of a redesign of the simd vectorization via wrappers, MPI improvements and a software redesign to allow memory-efficient execution with the production trunk to increase portability and extensibility. Two novel, memory-efficient OpenMP schemes for the linked cell-based force calculation are presented, which are able to retain Newton's third law optimization. Comparisons to well-optimized Verlet list-based codes, such as LAMMPS and GROMACS, demonstrate the viability of the linked cell-based approach. The present version of ls1 mardyn is used to run simulations on entire supercomputers, maximizing the number of sampled atoms. Compared to the preceding version of ls1 mardyn on the entire set of 9216 nodes of SuperMUC, Phase 1, 27% more atoms are simulated. Weak scaling performance is increased by up to 40% and strong scaling performance by up to more than 220%. On Hazel Hen, strong scaling efficiency of up to 81% and 189 billion molecule updates per second is attained, when scaling from 8 to 7168 nodes. Moreover, a total of 20 trillion atoms is simulated at up to 88% weak scaling efficiency running at up to 1.33 PFLOPS. This represents a fivefold increase in terms of the number of atoms simulated to date.

关键词： Molecular dynamics simd vectorization OpenMP MPI coloring memory-efficient SuperMUC Hazel Hen

来源：评论

学校读者我要写书评

暂无评论

Adaptive simd optimizations in particle-in-cell codes with fine-grain particle sorting

引用

COMPUTER PHYSICS COMMUNICATIONS 2019年 244卷 246-263页

作者： Beck, A. Derouillat, J. Lobet, M. Farjallah, A. Massimo, F. Zemzemi, I Perez, F. Vinci, T. Grech, M. Ecole Polytech Lab Leprince Ringuet CNRS IN2P3 F-91128 Palaiseau France Ecole Polytech LULI CNRS CEA F-91128 Palaiseau France UPMC Univ Paris 06 Sorbonne Univ CNRS LULI Pl Jussieu F-75252 Paris 05 France Univ Paris Saclay Univ Paris Sud UVSQ Maison SimulatCEACNRS F-91191 Gif Sur Yvette France Intel Corp Meudon France

Particle-In-Cell (PIC) codes are broadly applied to the kinetic simulation of plasmas, from laser matter interaction to astrophysics. Their heavy simulation cost can be mitigated by using the Single Instruction Multiple Data (simd) capability, or vectorization, now available on most architectures. This article details and discusses the vectorization strategy developed in the code SMILEI which takes advantage from an efficient, systematic, cell-based sorting of the particles. The PIC operators on particles (projection, push, deposition) have been optimized to benefit from large simd vectors on both recent and older architectures. The efficiency of these vectorized operations increases with the number of particles per cell (PPC), typically speeding up three-dimensional simulations by a factor 2 with 256 PPC. Although this implementation shows acceleration from as few as 8 PPC, it can be slower than the scalar version in domains containing fewer PPC as usually observed in vectorization attempts. This issue is overcome with an adaptive algorithm which switches locally between scalar (for few PPC) and vectorized operators (otherwise). The newly implemented methods are benchmarked on three different, large-scale simulations considering configurations frequently studied with PIC codes. (C) 2019 Elsevier B.V. All rights reserved.

关键词： Particle-in-cell (PIC) Sorting simd vectorization Plasma physics

来源：评论

学校读者我要写书评

暂无评论

Program Generation for Small-Scale Linear Algebra Applications 2018

Program Generation for Small-Scale Linear Algebra Applicatio...

引用

16th International Symposium on Code Generation and Optimization (CGO)

作者： Spampinato, Daniele G. Fabregat-Traver, Diego Bientinesi, Paolo Puschel, Markus Swiss Fed Inst Technol Dept Comp Sci Zurich Switzerland Rhein Westfal TH Aachen Aachen Inst Adv Study Computat Engn Sci Aachen Germany

ISBN: (纸本)9781450356176

We present SLINGEN, a program generation system for linear algebra. The input to SLINGEN is an application expressed mathematically in a linear-algebra-inspired language (LA) that we define. LA provides basic scalar/vector/matrix additions/multiplications and higher level operations including linear systems solvers, Cholesky and LU factorizations. The output of SLINGEN is performance-optimized single-source C code, optionally vectorized with intrinsics. The target of SLINGEN are small-scale computations on fixed-size operands, for which a straightforward implementation using optimized libraries (e.g., BLAS or LAPACK) is known to yield suboptimal performance (besides increasing code size and introducing dependencies), but which are crucial in control, signal processing, computer vision, and other domains. Internally, SLINGEN uses synthesis and DSL-based techniques to optimize at a high level of abstraction. We benchmark our program generator on three prototypical applications: the Kalman filter, Gaussian process regression, and an L1-analysis convex solver, as well as basic routines including Cholesky factorization and solvers for the continuous-time Lyapunov and Sylvester equations. The results show significant speed-ups compared to straightforward C with Intel icc and clang with a polyhedral optimizer, as well as library-based and template-based implementations.

关键词： Performance simd vectorization Rewriting

来源：评论

学校读者我要写书评

暂无评论

High-Performance Implementation of Matrix-Free Runge-Kutta Discontinuous Galerkin Method for Euler Equations 20

High-Performance Implementation of Matrix-Free Runge-Kutta D...

引用

20th IEEE International Conference on High Performance Computing and Communications (HPCC) / 16th IEEE International Conference on Smart City (SmartCity) / 4th IEEE International Conference on Data Science and Systems (DSS)

作者： Feng, Yongquan Yang, Wenjing Sun, Liaoyuan Lin, Zhipeng Zhang, Yongjun Natl Univ Def Technol Coll Comp State Key Lab High Performance Comp Changsha Hunan Peoples R China Natl Innovat Inst Def Technol Beijing Peoples R China

ISBN: (纸本)9781538666142

DG method is one of the mainstream high-order numerical discretization methods of CFD, and the RKDG which combines the Runge-Kutta explicit time stepping with DG space discretization plays an important role in the DG explicit numerical simulation. Even though the matrix-free has already been applied to improving the performance of the DG method, it has gained less attention relatively. In order to improve the RKDG's performance for Euler equations, we have implemented a matrix-free RKDG on the HopeFOAM, an open source CFD code. In this paper, we design and implement data structures to apply sum-factorization to the RKDG space discretization. During the simulation process, the arithmetic operations and memory cost decrease, which will improve the serial performance. For the data-level parallelism, we design the calculation strategy and vectorize the data to make it ideally suitable for the simd instructions, so that we make the most of the multiple processing elements of CPU. The experiment results show that matrix-free implementation can effectively speed up RKDG simulation on Euler equations and the performance improvement in three-dimensional space is much better than that in two-dimensions, whose maximum speed-up ratios are 26.287 and 8.278 respectively. Moreover, we test the parallel scalability of matrix-free RKDG implementation and the result indicates that matrix-free implementation does not lower the parallel speed-up ratio of HopeFOAM.

关键词： RKDG Euler Equations Sum-factorization simd vectorization

来源：评论

学校读者我要写书评

暂无评论

Distributed memory building blocks for massive biological sequence analysis

Distributed memory building blocks for massive biological se...

引用

作者： Pan, Tony C. Georgia Institute of Technology

学位级别：博士

K-mer indices and de Bruijn graphs are important data structures in bioinformatics with multiple applications ranging from foundational tasks such as error correction, alignment, and genome assembly, to knowledge discovery tasks including repeat detection and SNP identification. While advances in next generation sequencing technologies have dramatically reduced the cost and improved latency and throughput, few bioinformatics tools can efficiently process the data sets at the current generation rate of 1.8 terabases every 3 days. The volume and velocity with which sequencing data is generated necessitate efficient algorithms and implementation of k-mer indices and de Bruijn graphs, two central components in bioinformatic applications. Existing applications that utilize k-mer counting and de Bruijn graphs, however, tend to provide embedded, specialized implementations. The research presented here represents efforts toward the creation of the first reusable, flexible, and extensible distributed memory parallel libraries for k-mer indexing and de Bruijn graphs. These libraries are intended for simplifying the development of bioinformatics applications for distributed memory environments. For each library, our goals are to create a set of API that are simple to use, and provide optimized implementations based on efficient parallel algorithms. We designed algorithms that minimize communication volume and latency, and developed implementations with better cache utilization and simd vectorization. We developed Kmerind, a k-mer counting and indexing library based on distributed memory hash table and distributed sorted arrays, that provide efficient insert, find, count, and erase operations. For de Bruijn graphs, we developed Bruno by leveraging Kmerind functionalities to support parallel de Bruijn graph construction, chain compaction, error removal, and graph traversal and element query. Our performance evaluations showed that Kmerind is scalable and high performance. Kmerin

关键词： High performance computing Bioinformatics K-mer index K-mer counting De bruijn graph Next generation sequencing Parallel algorithms Distributed memory Distributed algorithms simd vectorization Cache friendly algorithms MPI

来源：评论

学校读者我要写书评

暂无评论

An efficient and portable simd algorithm for charge/current deposition in Particle-In-Cell codes

引用

COMPUTER PHYSICS COMMUNICATIONS 2017年 210卷 145-154页

作者： Vincenti, H. Lobet, M. Lehe, R. Sasanka, R. Vay, J. -L. Lawrence Berkeley Natl Lab 1 Cyclotron Rd Berkeley CA 94720 USA CEA Lasers Interact & Dynam Lab LIDyL Gif Sur Yvette France Intel Corp Hillsboro OR 97124 USA

In current computer architectures, data movement (from die to network) is by far the most energy consuming part of an algorithm (R,'20 pi/word on-die to 10,000 pi/word on the network). To increase memory locality at the hardware level and reduce energy consumption related to data movement, future exascale. computers tend to use many-core processors on each compute nodes that will have a reduced clock speed to allow for efficient cooling. To compensate for frequency decrease, machine vendors are making use of long simd instruction registers that are able to process multiple data with one arithmetic operator in one clock cycle. simd register length is expected to double every four years. As a consequence, Particle-In-Cell (PIC) codes will have to achieve good vectorization to fully take advantage of these upcoming architectures. In this paper, we present a new algorithm that allows for efficient and portable simd vectorization of current/charge deposition routines that are, along with the field gathering routines, among the most time consuming parts of the PIC algorithm. Our new algorithm uses a particular data structure that takes into account memory alignment constraints and avoids gather/scatter instructions that can significantly affect vectorization performances on current CPUs. The new algorithm was successfully implemented in the 3D skeleton PIC code PICSAR and tested on Haswell Xeon processors (AVX2-256 bits wide data registers). Results show a factor of x 2 to x2.5 speed-up in double precision for particle shape factor of orders 1-3. The new algorithm can be applied as is on future KNL (Knights Landing) architectures that will include AVX-512 instruction sets with 512 bits register lengths (8 doubles/16 singles). Program summary Program Title: vec_deposition Program Files doi: http://***/10.17632/nh77fv9k8c.1 Licensing provisions: BSD 3-Clause Programming language: Fortran 90 External routines/libraries: OpenMP > 4.0 Nature of problem: Exascale archite

关键词： Particle-In-Cell method OpenMP simd vectorization AVX2 AVX512 Tiling Cache reuse Many-core architectures

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：