Empirical Modeling: an Auto-tuning Method for Linear Algebra Routines on CPU plus Multi-GPU Platforms

Size: px
Start display at page:

Download "Empirical Modeling: an Auto-tuning Method for Linear Algebra Routines on CPU plus Multi-GPU Platforms"

Transcription

1 Empirical Modeling: an Auto-tuning Method for Linear Algebra Routines on CPU plus Multi-GPU Platforms Javier Cuenca Luis-Pedro García Domingo Giménez Francisco J. Herrera Scientific Computing and Parallel Programming

2 Motivation WHY: Scientific software must be optimized, by hand or automatically WHERE: Current hybrid computational systems: CPU plus several GPUs WHAT: Improving linear algebra routine performance a balanced assignation of the work to the processing units the best number of CPU threads HOW: Auto-tuning technique: execution time modeling WHEN: At installation time, mainly 2

3 Outline Routine: hybrid matrix multiplication kernel Auto-tuning method: empirical modeling Design phase Installation phase Execution phase Experimental results Conclusion and future work 3

4 Outline Routine: hybrid matrix multiplication kernel Auto-tuning method: empirical modeling Design phase Installation phase Execution phase Experimental results Conclusion and future work 4

5 Routine: hybrid matrix multiplication kernel C = aab + bc C = a (AB1 + AB2) + b(c1+c2) aab1 + bc1 to the GPU aab2 + bc2 to the CPU 5

6 Routine: hybrid matrix multiplication kernel C = aab + bc C = a (AB1+ +Abc+ABcpu) + b(c1+ +Cc+Ccpu) aabi + bci to the GPU i aabcpu c+ bccpu to the CPU 6

7 Outline Routine: hybrid matrix multiplication kernel Auto-tuning method: empirical modeling Design phase Installation phase Execution phase Experimental results Conclusion and future work 7

8 Auto-tuning method: empirical modeling Auto-tuning process. Three main phases: Design phase: building a Theoretical Model of the routine Texecution = function(n, System Parameters, Algorithmic Parameters) Installation phase: including computing and communication capacities of the current machine in the Model System Parameters Execution phase: deciding the values for a set of configurable parameters in the Model Algorithmic Parameters 8

9 Outline Routine: hybrid matrix multiplication kernel Auto-tuning method: empirical modeling Design phase Installation phase Execution phase Experimental results Conclusion and future work 9

10 Auto-tuning method: empirical modeling Design phase: building a theoretical model of the routine Computing time in each GPU i: n: dimension of the matrices Tcomp_gpu_i(n,n gpu_i ) = 2n 2 n gpu_i k comp_gpu_i n gpu_i : portion of matrix B assigned to GPU i (Algorithmic Parameter) k comp_gpu_i : average time to perform a basic operation (a double precision multiplication or addition) in GPU i (Computing System Parameter) The communication time CPU GPU i: T comm_gpu_i (n, n gpu_i ) = 3t s_i + (n 2 + 2nn gpu_i )t w_i t s_i and t w_i are Communication System Parameters t s_i : the average time for starting the communication t w_i : the average time for sending/receiving double precision number to/from the CPU Time to perform the work assigned to GPU i: T gpu_i (n, n gpu_i ) = T comp_gpu_i (n, n gpu_i ) + T comm_gpu_i (n, n gpu_i ) 10

11 Auto-tuning method: empirical modeling Design phase: building a theoretical model of the routine Computing time in the CPU: n: dimension of the matrices Tcomp_cpu(n,n cpu,t) = 2n 2 n cpu k comp_cpu_t n cpu : portion of matrix B assigned to the CPU (Algorithmic Parameter) t: number of CPU threads (Algorithmic Parameter) k comp_cpu_t : average time to perform a basic operation in the CPU with t threads (Computing System Parameter). Time to perform the work assigned to the CPU: T cpu (n, n cpu,t) = Tcomp_cpu(n,n cpu,t) 11

12 Auto-tuning method: empirical modeling Design phase: building a theoretical model of the routine A problem n r x n r to be solved Communications with GPUs overlap with CPU computing Total execution time the processing unit with the largest time: Ttotal(n r, n gpu_1,..., n gpu_c,n cpu,t) = max { Tgpu_1(n r, n gpu_1 ),..., Tgpu_c(n r, n gpu_c ), Tcpu(n r, n cpu, t) } 12

13 Outline Routine: hybrid matrix multiplication kernel Auto-tuning method: empirical modeling Design phase Installation phase Execution phase Experimental results Conclusion and future work 13

14 Auto-tuning method: empirical modeling Installation phase In the installation phase two main tasks: 1. Measuring System Parameter values: 2. Searching the best Algorithmic Parameter values for a fixed set of problem sizes 14

15 Auto-tuning method: empirical modeling Installation phase 1. Measuring System Parameter values: System Parameter values depend on the quantity of data In GPU i: For the Computing System Parameter, k comp_gpu_i : A benchmark applied to a set of local subproblems a set of execution times The set of execution times + Model a set of values for k comp_gpu_i For the Communication System Parameters, t s_i and t w_i A benchmark applied to a set of local subproblems a set of execution times The set of execution times + Model a set of values for t s_i and t w_i In CPU with t threads (t = 1,,number of cores): For the Computing System Parameter, k comp_cpu_t : A benchmark applied to a set of local subproblems a set of execution times The set of execution times + Model a set of values for k comp_gpu_t 15

16 Auto-tuning method: empirical modeling Installation phase 2. Searching the best Algorithmic Parameter values for a fixed set of problem sizes Global_ Problem_ Size_Set: a set of data sizes for the whole problem (System Manager) For each problem size from Global_ Problem_ Size_Set A complete searching process of the best Algorithmic Parameter values (according to the Model with System Parameter values). The selected Algorithmic Parameter values are added to Global_Best_AP_Installation_Set 16

17 Auto-tuning method: empirical modeling Installation phase 2. Searching the best Algorithmic Parameter values for a fixed set of problem sizes Almost the whole installation time is devoted to this second part This time depends on: The searching space dimensions (number of GPUs and the number of CPU threads) The Distribution Grain Size, DGS (decided by System Manager) DGS value conditions: The installation time The goodness of the future taken decisions 17

18 Outline Routine: hybrid matrix multiplication kernel Auto-tuning method: empirical modeling Design phase Installation phase Execution phase Experimental results Conclusion and future work 18

19 Auto-tuning method: empirical modeling Execution phase At runtime, a problem n r x n r to be solved The Algorithmic Parameter values to use: those in Global_Best_ AP_Installation_Set for the n closest to n r None significant overload in the total execution time 19

20 Outline Routine: hybrid matrix multiplication kernel Auto-tuning method: empirical modeling Design phase Installation phase Execution phase Experimental results Conclusion and future work 20

21 Experimental results Platform: 12CtwoC2075fourGTX590 A shared-memory system with: 2 hexa-cores (12 cores) Intel Xeon E GPU devices Nvidia Fermi Tesla C2075 with: 5375 MBytes of Global Memory 448 cores: 14 Streaming Multiprocessors, 32 Streaming Processors per Multiprocessor 4 GPU Nvidia GeForce GTX 590 with: 1536 MBytes in Global Memory 512 cores :16 Streaming Multiprocessors, 32 Streaming Processors per Multiprocessor The different system configurations: 12CxC2075yGTX590, to indicate: a CPU with 12 cores x Fermi Tesla C2075 y GeForce GTX

22 Time (seconds) Time (seconds) Time (seconds) Time (seconds) Experimental results Platform: 12ConeC2075twoGTX590 model vs. experimental execution time. CPU threads=10. 0,150 n=1000 1,000 0,800 n=2000 0,100 0,600 0,050 0, work distributions modeling time execution time 0,400 0,200 0, work distributions modeling time execution time n=3000 n=4000 4,000 8,000 3,000 6,000 2,000 modeling time 4,000 1,000 execution time 2,000 0,000 0, work distributions work distributions modeling time execution time 22

23 Experimental results Platform: 12CtwoC2075fourGTX590 Auto-tuning vs. optimal. CPU threads=10, DSG=10% of n 23

24 Experimental results Platform: 12CtwoC2075fourGTX590 GFLOPS and installation time with different Distribution Grain Sizes (DGS) 24

25 Experimental results Different Platforms GFLOPS with different work distribution methods. 25

26 Experimental results Different Platforms GFLOPS with different work distribution methods. 26

27 Outline Routine: hybrid matrix multiplication kernel Auto-tuning method: empirical modeling Design phase Installation phase Execution phase Experimental results Conclusion and future work 27

28 Conclusion and future work Autotuning methodology: empirical modeling. Adaptation to multicore+multigpu systems. To obtain: Balanced distributions of the work Best number of CPU threads Linear algebra routines. Example: a matrix-matrix multiplication Satisfactory results for these complex, heterogeneous systems are reported More experiments: in more systems with larger numbers of coprocessors of different architectures with other linear algebra routines More complex systems should be considered, for example multicore+multigpu+multimic 28

Improving Linear Algebra Computation on NUMA platforms through auto-tuned tuned nested parallelism

Improving Linear Algebra Computation on NUMA platforms through auto-tuned tuned nested parallelism Improving Linear Algebra Computation on NUMA platforms through auto-tuned tuned nested parallelism Javier Cuenca, Luis P. García, Domingo Giménez Parallel Computing Group University of Murcia, SPAIN parallelum

More information

MAGMA a New Generation of Linear Algebra Libraries for GPU and Multicore Architectures

MAGMA a New Generation of Linear Algebra Libraries for GPU and Multicore Architectures MAGMA a New Generation of Linear Algebra Libraries for GPU and Multicore Architectures Stan Tomov Innovative Computing Laboratory University of Tennessee, Knoxville OLCF Seminar Series, ORNL June 16, 2010

More information

High performance 2D Discrete Fourier Transform on Heterogeneous Platforms. Shrenik Lad, IIIT Hyderabad Advisor : Dr. Kishore Kothapalli

High performance 2D Discrete Fourier Transform on Heterogeneous Platforms. Shrenik Lad, IIIT Hyderabad Advisor : Dr. Kishore Kothapalli High performance 2D Discrete Fourier Transform on Heterogeneous Platforms Shrenik Lad, IIIT Hyderabad Advisor : Dr. Kishore Kothapalli Motivation Fourier Transform widely used in Physics, Astronomy, Engineering

More information

Data Partitioning on Heterogeneous Multicore and Multi-GPU systems Using Functional Performance Models of Data-Parallel Applictions

Data Partitioning on Heterogeneous Multicore and Multi-GPU systems Using Functional Performance Models of Data-Parallel Applictions Data Partitioning on Heterogeneous Multicore and Multi-GPU systems Using Functional Performance Models of Data-Parallel Applictions Ziming Zhong Vladimir Rychkov Alexey Lastovetsky Heterogeneous Computing

More information

International Conference on Computational Science (ICCS 2017)

International Conference on Computational Science (ICCS 2017) International Conference on Computational Science (ICCS 2017) Exploiting Hybrid Parallelism in the Kinematic Analysis of Multibody Systems Based on Group Equations G. Bernabé, J. C. Cano, J. Cuenca, A.

More information

Solving Dense Linear Systems on Graphics Processors

Solving Dense Linear Systems on Graphics Processors Solving Dense Linear Systems on Graphics Processors Sergio Barrachina Maribel Castillo Francisco Igual Rafael Mayo Enrique S. Quintana-Ortí High Performance Computing & Architectures Group Universidad

More information

NVIDIA s Compute Unified Device Architecture (CUDA)

NVIDIA s Compute Unified Device Architecture (CUDA) NVIDIA s Compute Unified Device Architecture (CUDA) Mike Bailey mjb@cs.oregonstate.edu Reaching the Promised Land NVIDIA GPUs CUDA Knights Corner Speed Intel CPUs General Programmability 1 History of GPU

More information

NVIDIA s Compute Unified Device Architecture (CUDA)

NVIDIA s Compute Unified Device Architecture (CUDA) NVIDIA s Compute Unified Device Architecture (CUDA) Mike Bailey mjb@cs.oregonstate.edu Reaching the Promised Land NVIDIA GPUs CUDA Knights Corner Speed Intel CPUs General Programmability History of GPU

More information

Applications of Berkeley s Dwarfs on Nvidia GPUs

Applications of Berkeley s Dwarfs on Nvidia GPUs Applications of Berkeley s Dwarfs on Nvidia GPUs Seminar: Topics in High-Performance and Scientific Computing Team N2: Yang Zhang, Haiqing Wang 05.02.2015 Overview CUDA The Dwarfs Dynamic Programming Sparse

More information

Parallelism on Hybrid Metaheuristics for Vector Autoregression Models

Parallelism on Hybrid Metaheuristics for Vector Autoregression Models Parallelism on Hybrid Metaheuristics for Vector Autoregression Models Alfonso L. Castaño, Javier Cuenca, José-Matías Cutillas-Lozano, Domingo Giménez Scientific Computing and Parallel Programming Group,

More information

Coordinating More Than 3 Million CUDA Threads for Social Network Analysis. Adam McLaughlin

Coordinating More Than 3 Million CUDA Threads for Social Network Analysis. Adam McLaughlin Coordinating More Than 3 Million CUDA Threads for Social Network Analysis Adam McLaughlin Applications of interest Computational biology Social network analysis Urban planning Epidemiology Hardware verification

More information

QR Decomposition on GPUs

QR Decomposition on GPUs QR Decomposition QR Algorithms Block Householder QR Andrew Kerr* 1 Dan Campbell 1 Mark Richards 2 1 Georgia Tech Research Institute 2 School of Electrical and Computer Engineering Georgia Institute of

More information

Performance Analysis of Memory Transfers and GEMM Subroutines on NVIDIA TESLA GPU Cluster

Performance Analysis of Memory Transfers and GEMM Subroutines on NVIDIA TESLA GPU Cluster Performance Analysis of Memory Transfers and GEMM Subroutines on NVIDIA TESLA GPU Cluster Veerendra Allada, Troy Benjegerdes Electrical and Computer Engineering, Ames Laboratory Iowa State University &

More information

Document downloaded from: This paper must be cited as:

Document downloaded from:   This paper must be cited as: Document downloaded from: http://hdl.handle.net/1/ This paper must be cited as: Cámara, J.; Cuenca, J.; Giménez, D.; García, LP.; Vidal Maciá, AM. (01). Empirical Installation of Linear Algebra Shared-Memory

More information

CMSC 714 Lecture 6 MPI vs. OpenMP and OpenACC. Guest Lecturer: Sukhyun Song (original slides by Alan Sussman)

CMSC 714 Lecture 6 MPI vs. OpenMP and OpenACC. Guest Lecturer: Sukhyun Song (original slides by Alan Sussman) CMSC 714 Lecture 6 MPI vs. OpenMP and OpenACC Guest Lecturer: Sukhyun Song (original slides by Alan Sussman) Parallel Programming with Message Passing and Directives 2 MPI + OpenMP Some applications can

More information

A proposal for autotuning linear algebra routines on multicore platforms

A proposal for autotuning linear algebra routines on multicore platforms Procedia Computer Science 001 (2010) (2012) 1 9 515 523 Procedia Computer Science www.elsevier.com/locate/procedia International Conference on Computational Science, ICCS 2010 A proposal for autotuning

More information

Auto-Generation and Auto-Tuning of 3D Stencil Codes on GPU Clusters

Auto-Generation and Auto-Tuning of 3D Stencil Codes on GPU Clusters Auto-Generation and Auto-Tuning of 3D Stencil s on GPU Clusters Yongpeng Zhang, Frank Mueller North Carolina State University CGO 2012 Outline Motivation DSL front-end and Benchmarks Framework Experimental

More information

Introduction to CUDA Algoritmi e Calcolo Parallelo. Daniele Loiacono

Introduction to CUDA Algoritmi e Calcolo Parallelo. Daniele Loiacono Introduction to CUDA Algoritmi e Calcolo Parallelo References q This set of slides is mainly based on: " CUDA Technical Training, Dr. Antonino Tumeo, Pacific Northwest National Laboratory " Slide of Applied

More information

Finite Element Integration and Assembly on Modern Multi and Many-core Processors

Finite Element Integration and Assembly on Modern Multi and Many-core Processors Finite Element Integration and Assembly on Modern Multi and Many-core Processors Krzysztof Banaś, Jan Bielański, Kazimierz Chłoń AGH University of Science and Technology, Mickiewicza 30, 30-059 Kraków,

More information

CME 213 S PRING Eric Darve

CME 213 S PRING Eric Darve CME 213 S PRING 2017 Eric Darve Summary of previous lectures Pthreads: low-level multi-threaded programming OpenMP: simplified interface based on #pragma, adapted to scientific computing OpenMP for and

More information

Parallel FFT Program Optimizations on Heterogeneous Computers

Parallel FFT Program Optimizations on Heterogeneous Computers Parallel FFT Program Optimizations on Heterogeneous Computers Shuo Chen, Xiaoming Li Department of Electrical and Computer Engineering University of Delaware, Newark, DE 19716 Outline Part I: A Hybrid

More information

Modelling and optimisation of scientific software for multicore platforms

Modelling and optimisation of scientific software for multicore platforms Modelling and optimisation of scientific software for multicore platforms Domingo Giménez... and the list of collaborators within the presentation Group page: http://www.um.es/pcgum Presentations: http://dis.um.es/%7edomingo/investigacion.html

More information

Solving Dense Linear Systems on Platforms with Multiple Hardware Accelerators

Solving Dense Linear Systems on Platforms with Multiple Hardware Accelerators Solving Dense Linear Systems on Platforms with Multiple Hardware Accelerators Francisco D. Igual Enrique S. Quintana-Ortí Gregorio Quintana-Ortí Universidad Jaime I de Castellón (Spain) Robert A. van de

More information

GPU ACCELERATION OF WSMP (WATSON SPARSE MATRIX PACKAGE)

GPU ACCELERATION OF WSMP (WATSON SPARSE MATRIX PACKAGE) GPU ACCELERATION OF WSMP (WATSON SPARSE MATRIX PACKAGE) NATALIA GIMELSHEIN ANSHUL GUPTA STEVE RENNICH SEID KORIC NVIDIA IBM NVIDIA NCSA WATSON SPARSE MATRIX PACKAGE (WSMP) Cholesky, LDL T, LU factorization

More information

Use of Synthetic Benchmarks for Machine- Learning-based Performance Auto-tuning

Use of Synthetic Benchmarks for Machine- Learning-based Performance Auto-tuning Use of Synthetic Benchmarks for Machine- Learning-based Performance Auto-tuning Tianyi David Han and Tarek S. Abdelrahman The Edward S. Rogers Department of Electrical and Computer Engineering University

More information

High Performance Computing on GPUs using NVIDIA CUDA

High Performance Computing on GPUs using NVIDIA CUDA High Performance Computing on GPUs using NVIDIA CUDA Slides include some material from GPGPU tutorial at SIGGRAPH2007: http://www.gpgpu.org/s2007 1 Outline Motivation Stream programming Simplified HW and

More information

MAGMA. Matrix Algebra on GPU and Multicore Architectures

MAGMA. Matrix Algebra on GPU and Multicore Architectures MAGMA Matrix Algebra on GPU and Multicore Architectures Innovative Computing Laboratory Electrical Engineering and Computer Science University of Tennessee Piotr Luszczek (presenter) web.eecs.utk.edu/~luszczek/conf/

More information

Accelerating GPU Kernels for Dense Linear Algebra

Accelerating GPU Kernels for Dense Linear Algebra Accelerating GPU Kernels for Dense Linear Algebra Rajib Nath, Stan Tomov, and Jack Dongarra Innovative Computing Lab University of Tennessee, Knoxville July 9, 21 xgemm performance of CUBLAS-2.3 on GTX28

More information

University of Bielefeld

University of Bielefeld Geistes-, Natur-, Sozial- und Technikwissenschaften gemeinsam unter einem Dach Introduction to GPU Programming using CUDA Olaf Kaczmarek University of Bielefeld STRONGnet Summerschool 2011 ZIF Bielefeld

More information

G P G P U : H I G H - P E R F O R M A N C E C O M P U T I N G

G P G P U : H I G H - P E R F O R M A N C E C O M P U T I N G Joined Advanced Student School (JASS) 2009 March 29 - April 7, 2009 St. Petersburg, Russia G P G P U : H I G H - P E R F O R M A N C E C O M P U T I N G Dmitry Puzyrev St. Petersburg State University Faculty

More information

Carlos Reaño, Javier Prades and Federico Silla Technical University of Valencia (Spain)

Carlos Reaño, Javier Prades and Federico Silla Technical University of Valencia (Spain) Carlos Reaño, Javier Prades and Federico Silla Technical University of Valencia (Spain) 4th IEEE International Workshop of High-Performance Interconnection Networks in the Exascale and Big-Data Era (HiPINEB

More information

Introduction to CUDA

Introduction to CUDA Introduction to CUDA Overview HW computational power Graphics API vs. CUDA CUDA glossary Memory model, HW implementation, execution Performance guidelines CUDA compiler C/C++ Language extensions Limitations

More information

Mathematical computations with GPUs

Mathematical computations with GPUs Master Educational Program Information technology in applications Mathematical computations with GPUs GPU architecture Alexey A. Romanenko arom@ccfit.nsu.ru Novosibirsk State University GPU Graphical Processing

More information

High-Order Finite-Element Earthquake Modeling on very Large Clusters of CPUs or GPUs

High-Order Finite-Element Earthquake Modeling on very Large Clusters of CPUs or GPUs High-Order Finite-Element Earthquake Modeling on very Large Clusters of CPUs or GPUs Gordon Erlebacher Department of Scientific Computing Sept. 28, 2012 with Dimitri Komatitsch (Pau,France) David Michea

More information

Auto-tunable GPU BLAS

Auto-tunable GPU BLAS Auto-tunable GPU BLAS Jarle Erdal Steinsland Master of Science in Computer Science Submission date: June 2011 Supervisor: Anne Cathrine Elster, IDI Norwegian University of Science and Technology Department

More information

GPGPUs in HPC. VILLE TIMONEN Åbo Akademi University CSC

GPGPUs in HPC. VILLE TIMONEN Åbo Akademi University CSC GPGPUs in HPC VILLE TIMONEN Åbo Akademi University 2.11.2010 @ CSC Content Background How do GPUs pull off higher throughput Typical architecture Current situation & the future GPGPU languages A tale of

More information

A Large-Scale Cross-Architecture Evaluation of Thread-Coarsening. Alberto Magni, Christophe Dubach, Michael O'Boyle

A Large-Scale Cross-Architecture Evaluation of Thread-Coarsening. Alberto Magni, Christophe Dubach, Michael O'Boyle A Large-Scale Cross-Architecture Evaluation of Thread-Coarsening Alberto Magni, Christophe Dubach, Michael O'Boyle Introduction Wide adoption of GPGPU for HPC Many GPU devices from many of vendors AMD

More information

CUDA PROGRAMMING MODEL Chaithanya Gadiyam Swapnil S Jadhav

CUDA PROGRAMMING MODEL Chaithanya Gadiyam Swapnil S Jadhav CUDA PROGRAMMING MODEL Chaithanya Gadiyam Swapnil S Jadhav CMPE655 - Multiple Processor Systems Fall 2015 Rochester Institute of Technology Contents What is GPGPU? What s the need? CUDA-Capable GPU Architecture

More information

On Level Scheduling for Incomplete LU Factorization Preconditioners on Accelerators

On Level Scheduling for Incomplete LU Factorization Preconditioners on Accelerators On Level Scheduling for Incomplete LU Factorization Preconditioners on Accelerators Karl Rupp, Barry Smith rupp@mcs.anl.gov Mathematics and Computer Science Division Argonne National Laboratory FEMTEC

More information

CUDA and OpenCL Implementations of 3D CT Reconstruction for Biomedical Imaging

CUDA and OpenCL Implementations of 3D CT Reconstruction for Biomedical Imaging CUDA and OpenCL Implementations of 3D CT Reconstruction for Biomedical Imaging Saoni Mukherjee, Nicholas Moore, James Brock and Miriam Leeser September 12, 2012 1 Outline Introduction to CT Scan, 3D reconstruction

More information

Performance Diagnosis for Hybrid CPU/GPU Environments

Performance Diagnosis for Hybrid CPU/GPU Environments Performance Diagnosis for Hybrid CPU/GPU Environments Michael M. Smith and Karen L. Karavanic Computer Science Department Portland State University Performance Diagnosis for Hybrid CPU/GPU Environments

More information

How to perform HPL on CPU&GPU clusters. Dr.sc. Draško Tomić

How to perform HPL on CPU&GPU clusters. Dr.sc. Draško Tomić How to perform HPL on CPU&GPU clusters Dr.sc. Draško Tomić email: drasko.tomic@hp.com Forecasting is not so easy, HPL benchmarking could be even more difficult Agenda TOP500 GPU trends Some basics about

More information

Introduction to Parallel and Distributed Computing. Linh B. Ngo CPSC 3620

Introduction to Parallel and Distributed Computing. Linh B. Ngo CPSC 3620 Introduction to Parallel and Distributed Computing Linh B. Ngo CPSC 3620 Overview: What is Parallel Computing To be run using multiple processors A problem is broken into discrete parts that can be solved

More information

A MATLAB Interface to the GPU

A MATLAB Interface to the GPU Introduction Results, conclusions and further work References Department of Informatics Faculty of Mathematics and Natural Sciences University of Oslo June 2007 Introduction Results, conclusions and further

More information

GPU Acceleration of the Longwave Rapid Radiative Transfer Model in WRF using CUDA Fortran. G. Ruetsch, M. Fatica, E. Phillips, N.

GPU Acceleration of the Longwave Rapid Radiative Transfer Model in WRF using CUDA Fortran. G. Ruetsch, M. Fatica, E. Phillips, N. GPU Acceleration of the Longwave Rapid Radiative Transfer Model in WRF using CUDA Fortran G. Ruetsch, M. Fatica, E. Phillips, N. Juffa Outline WRF and RRTM Previous Work CUDA Fortran Features RRTM in CUDA

More information

CUDA Architecture & Programming Model

CUDA Architecture & Programming Model CUDA Architecture & Programming Model Course on Multi-core Architectures & Programming Oliver Taubmann May 9, 2012 Outline Introduction Architecture Generation Fermi A Brief Look Back At Tesla What s New

More information

Introduction to CUDA Algoritmi e Calcolo Parallelo. Daniele Loiacono

Introduction to CUDA Algoritmi e Calcolo Parallelo. Daniele Loiacono Introduction to CUDA Algoritmi e Calcolo Parallelo References This set of slides is mainly based on: CUDA Technical Training, Dr. Antonino Tumeo, Pacific Northwest National Laboratory Slide of Applied

More information

X10 specific Optimization of CPU GPU Data transfer with Pinned Memory Management

X10 specific Optimization of CPU GPU Data transfer with Pinned Memory Management X10 specific Optimization of CPU GPU Data transfer with Pinned Memory Management Hideyuki Shamoto, Tatsuhiro Chiba, Mikio Takeuchi Tokyo Institute of Technology IBM Research Tokyo Programming for large

More information

GPU Programming. Lecture 2: CUDA C Basics. Miaoqing Huang University of Arkansas 1 / 34

GPU Programming. Lecture 2: CUDA C Basics. Miaoqing Huang University of Arkansas 1 / 34 1 / 34 GPU Programming Lecture 2: CUDA C Basics Miaoqing Huang University of Arkansas 2 / 34 Outline Evolvements of NVIDIA GPU CUDA Basic Detailed Steps Device Memories and Data Transfer Kernel Functions

More information

Adaptive-Mesh-Refinement Hydrodynamic GPU Computation in Astrophysics

Adaptive-Mesh-Refinement Hydrodynamic GPU Computation in Astrophysics Adaptive-Mesh-Refinement Hydrodynamic GPU Computation in Astrophysics H. Y. Schive ( 薛熙于 ) Graduate Institute of Physics, National Taiwan University Leung Center for Cosmology and Particle Astrophysics

More information

Scheduling FFT Computation on SMP and Multicore Systems Ayaz Ali, Lennart Johnsson & Jaspal Subhlok

Scheduling FFT Computation on SMP and Multicore Systems Ayaz Ali, Lennart Johnsson & Jaspal Subhlok Scheduling FFT Computation on SMP and Multicore Systems Ayaz Ali, Lennart Johnsson & Jaspal Subhlok Texas Learning and Computation Center Department of Computer Science University of Houston Outline Motivation

More information

CUDA GPGPU Workshop 2012

CUDA GPGPU Workshop 2012 CUDA GPGPU Workshop 2012 Parallel Programming: C thread, Open MP, and Open MPI Presenter: Nasrin Sultana Wichita State University 07/10/2012 Parallel Programming: Open MP, MPI, Open MPI & CUDA Outline

More information

An Extension of the StarSs Programming Model for Platforms with Multiple GPUs

An Extension of the StarSs Programming Model for Platforms with Multiple GPUs An Extension of the StarSs Programming Model for Platforms with Multiple GPUs Eduard Ayguadé 2 Rosa M. Badia 2 Francisco Igual 1 Jesús Labarta 2 Rafael Mayo 1 Enrique S. Quintana-Ortí 1 1 Departamento

More information

Hierarchical DAG Scheduling for Hybrid Distributed Systems

Hierarchical DAG Scheduling for Hybrid Distributed Systems June 16, 2015 Hierarchical DAG Scheduling for Hybrid Distributed Systems Wei Wu, Aurelien Bouteiller, George Bosilca, Mathieu Faverge, Jack Dongarra IPDPS 2015 Outline! Introduction & Motivation! Hierarchical

More information

n N c CIni.o ewsrg.au

n N c CIni.o ewsrg.au @NCInews NCI and Raijin National Computational Infrastructure 2 Our Partners General purpose, highly parallel processors High FLOPs/watt and FLOPs/$ Unit of execution Kernel Separate memory subsystem GPGPU

More information

Lecture 15: Introduction to GPU programming. Lecture 15: Introduction to GPU programming p. 1

Lecture 15: Introduction to GPU programming. Lecture 15: Introduction to GPU programming p. 1 Lecture 15: Introduction to GPU programming Lecture 15: Introduction to GPU programming p. 1 Overview Hardware features of GPGPU Principles of GPU programming A good reference: David B. Kirk and Wen-mei

More information

Massively Parallel Architectures

Massively Parallel Architectures Massively Parallel Architectures A Take on Cell Processor and GPU programming Joel Falcou - LRI joel.falcou@lri.fr Bat. 490 - Bureau 104 20 janvier 2009 Motivation The CELL processor Harder,Better,Faster,Stronger

More information

Large scale Imaging on Current Many- Core Platforms

Large scale Imaging on Current Many- Core Platforms Large scale Imaging on Current Many- Core Platforms SIAM Conf. on Imaging Science 2012 May 20, 2012 Dr. Harald Köstler Chair for System Simulation Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen,

More information

Study and implementation of computational methods for Differential Equations in heterogeneous systems. Asimina Vouronikoy - Eleni Zisiou

Study and implementation of computational methods for Differential Equations in heterogeneous systems. Asimina Vouronikoy - Eleni Zisiou Study and implementation of computational methods for Differential Equations in heterogeneous systems Asimina Vouronikoy - Eleni Zisiou Outline Introduction Review of related work Cyclic Reduction Algorithm

More information

Linear Algebra libraries in Debian. DebConf 10 New York 05/08/2010 Sylvestre

Linear Algebra libraries in Debian. DebConf 10 New York 05/08/2010 Sylvestre Linear Algebra libraries in Debian Who I am? Core developer of Scilab (daily job) Debian Developer Involved in Debian mainly in Science and Java aspects sylvestre.ledru@scilab.org / sylvestre@debian.org

More information

Accelerating image registration on GPUs

Accelerating image registration on GPUs Accelerating image registration on GPUs Harald Köstler, Sunil Ramgopal Tatavarty SIAM Conference on Imaging Science (IS10) 13.4.2010 Contents Motivation: Image registration with FAIR GPU Programming Combining

More information

CS8803SC Software and Hardware Cooperative Computing GPGPU. Prof. Hyesoon Kim School of Computer Science Georgia Institute of Technology

CS8803SC Software and Hardware Cooperative Computing GPGPU. Prof. Hyesoon Kim School of Computer Science Georgia Institute of Technology CS8803SC Software and Hardware Cooperative Computing GPGPU Prof. Hyesoon Kim School of Computer Science Georgia Institute of Technology Why GPU? A quiet revolution and potential build-up Calculation: 367

More information

The Uintah Framework: A Unified Heterogeneous Task Scheduling and Runtime System

The Uintah Framework: A Unified Heterogeneous Task Scheduling and Runtime System The Uintah Framework: A Unified Heterogeneous Task Scheduling and Runtime System Alan Humphrey, Qingyu Meng, Martin Berzins Scientific Computing and Imaging Institute & University of Utah I. Uintah Overview

More information

Parallelization of the QR Decomposition with Column Pivoting Using Column Cyclic Distribution on Multicore and GPU Processors

Parallelization of the QR Decomposition with Column Pivoting Using Column Cyclic Distribution on Multicore and GPU Processors Parallelization of the QR Decomposition with Column Pivoting Using Column Cyclic Distribution on Multicore and GPU Processors Andrés Tomás 1, Zhaojun Bai 1, and Vicente Hernández 2 1 Department of Computer

More information

Presenting: Comparing the Power and Performance of Intel's SCC to State-of-the-Art CPUs and GPUs

Presenting: Comparing the Power and Performance of Intel's SCC to State-of-the-Art CPUs and GPUs Presenting: Comparing the Power and Performance of Intel's SCC to State-of-the-Art CPUs and GPUs A paper comparing modern architectures Joakim Skarding Christian Chavez Motivation Continue scaling of performance

More information

arxiv: v1 [physics.comp-ph] 4 Nov 2013

arxiv: v1 [physics.comp-ph] 4 Nov 2013 arxiv:1311.0590v1 [physics.comp-ph] 4 Nov 2013 Performance of Kepler GTX Titan GPUs and Xeon Phi System, Weonjong Lee, and Jeonghwan Pak Lattice Gauge Theory Research Center, CTP, and FPRD, Department

More information

CS GPU and GPGPU Programming Lecture 8+9: GPU Architecture 7+8. Markus Hadwiger, KAUST

CS GPU and GPGPU Programming Lecture 8+9: GPU Architecture 7+8. Markus Hadwiger, KAUST CS 380 - GPU and GPGPU Programming Lecture 8+9: GPU Architecture 7+8 Markus Hadwiger, KAUST Reading Assignment #5 (until March 12) Read (required): Programming Massively Parallel Processors book, Chapter

More information

A GPU Implementation of Tiled Belief Propagation on Markov Random Fields. Hassan Eslami Theodoros Kasampalis Maria Kotsifakou

A GPU Implementation of Tiled Belief Propagation on Markov Random Fields. Hassan Eslami Theodoros Kasampalis Maria Kotsifakou A GPU Implementation of Tiled Belief Propagation on Markov Random Fields Hassan Eslami Theodoros Kasampalis Maria Kotsifakou BP-M AND TILED-BP 2 BP-M 3 Tiled BP T 0 T 1 T 2 T 3 T 4 T 5 T 6 T 7 T 8 4 Tiled

More information

A TALENTED CPU-TO-GPU MEMORY MAPPING TECHNIQUE

A TALENTED CPU-TO-GPU MEMORY MAPPING TECHNIQUE A TALENTED CPU-TO-GPU MEMORY MAPPING TECHNIQUE Abu Asaduzzaman, Deepthi Gummadi, and Chok M. Yip Department of Electrical Engineering and Computer Science Wichita State University Wichita, Kansas, USA

More information

Performance auto-tuning of rectangular matrix-vector multiplication: how to outperform CUBLAS

Performance auto-tuning of rectangular matrix-vector multiplication: how to outperform CUBLAS Downloaded from orbit.dtu.dk on: Dec 23, 2018 Performance auto-tuning of rectangular matrix-vector multiplication: how to outperform CUBLAS Sørensen, Hans Henrik Brandenborg Publication date: 2010 Document

More information

MAGMA: a New Generation

MAGMA: a New Generation 1.3 MAGMA: a New Generation of Linear Algebra Libraries for GPU and Multicore Architectures Jack Dongarra T. Dong, M. Gates, A. Haidar, S. Tomov, and I. Yamazaki University of Tennessee, Knoxville Release

More information

OpenCL implementation of PSO: a comparison between multi-core CPU and GPU performances

OpenCL implementation of PSO: a comparison between multi-core CPU and GPU performances OpenCL implementation of PSO: a comparison between multi-core CPU and GPU performances Stefano Cagnoni 1, Alessandro Bacchini 1,2, Luca Mussi 1 1 Dept. of Information Engineering, University of Parma,

More information

GAMER : a GPU-accelerated Adaptive-MEsh-Refinement Code for Astrophysics GPU 與自適性網格於天文模擬之應用與效能

GAMER : a GPU-accelerated Adaptive-MEsh-Refinement Code for Astrophysics GPU 與自適性網格於天文模擬之應用與效能 GAMER : a GPU-accelerated Adaptive-MEsh-Refinement Code for Astrophysics GPU 與自適性網格於天文模擬之應用與效能 Hsi-Yu Schive ( 薛熙于 ), Tzihong Chiueh ( 闕志鴻 ), Yu-Chih Tsai ( 蔡御之 ), Ui-Han Zhang ( 張瑋瀚 ) Graduate Institute

More information

Automated Finite Element Computations in the FEniCS Framework using GPUs

Automated Finite Element Computations in the FEniCS Framework using GPUs Automated Finite Element Computations in the FEniCS Framework using GPUs Florian Rathgeber (f.rathgeber10@imperial.ac.uk) Advanced Modelling and Computation Group (AMCG) Department of Earth Science & Engineering

More information

J. Blair Perot. Ali Khajeh-Saeed. Software Engineer CD-adapco. Mechanical Engineering UMASS, Amherst

J. Blair Perot. Ali Khajeh-Saeed. Software Engineer CD-adapco. Mechanical Engineering UMASS, Amherst Ali Khajeh-Saeed Software Engineer CD-adapco J. Blair Perot Mechanical Engineering UMASS, Amherst Supercomputers Optimization Stream Benchmark Stag++ (3D Incompressible Flow Code) Matrix Multiply Function

More information

CUDA Accelerated Linpack on Clusters. E. Phillips, NVIDIA Corporation

CUDA Accelerated Linpack on Clusters. E. Phillips, NVIDIA Corporation CUDA Accelerated Linpack on Clusters E. Phillips, NVIDIA Corporation Outline Linpack benchmark CUDA Acceleration Strategy Fermi DGEMM Optimization / Performance Linpack Results Conclusions LINPACK Benchmark

More information

Technical Report Performance Analysis of CULA on different NVIDIA GPU Architectures. Prateek Gupta

Technical Report Performance Analysis of CULA on different NVIDIA GPU Architectures. Prateek Gupta Technical Report 2014-02 Performance Analysis of CULA on different NVIDIA GPU Architectures Prateek Gupta May 20, 2014 1 Spring 2014: Performance Analysis of CULA on different NVIDIA GPU Architectures

More information

Automatic Development of Linear Algebra Libraries for the Tesla Series

Automatic Development of Linear Algebra Libraries for the Tesla Series Automatic Development of Linear Algebra Libraries for the Tesla Series Enrique S. Quintana-Ortí quintana@icc.uji.es Universidad Jaime I de Castellón (Spain) Dense Linear Algebra Major problems: Source

More information

CS 179: GPU Computing LECTURE 4: GPU MEMORY SYSTEMS

CS 179: GPU Computing LECTURE 4: GPU MEMORY SYSTEMS CS 179: GPU Computing LECTURE 4: GPU MEMORY SYSTEMS 1 Last time Each block is assigned to and executed on a single streaming multiprocessor (SM). Threads execute in groups of 32 called warps. Threads in

More information

Shadowfax: Scaling in Heterogeneous Cluster Systems via GPGPU Assemblies

Shadowfax: Scaling in Heterogeneous Cluster Systems via GPGPU Assemblies Shadowfax: Scaling in Heterogeneous Cluster Systems via GPGPU Assemblies Alexander Merritt, Vishakha Gupta, Abhishek Verma, Ada Gavrilovska, Karsten Schwan {merritt.alex,abhishek.verma}@gatech.edu {vishakha,ada,schwan}@cc.gtaech.edu

More information

"On the Capability and Achievable Performance of FPGAs for HPC Applications"

On the Capability and Achievable Performance of FPGAs for HPC Applications "On the Capability and Achievable Performance of FPGAs for HPC Applications" Wim Vanderbauwhede School of Computing Science, University of Glasgow, UK Or in other words "How Fast Can Those FPGA Thingies

More information

Graphics Processor Acceleration and YOU

Graphics Processor Acceleration and YOU Graphics Processor Acceleration and YOU James Phillips Research/gpu/ Goals of Lecture After this talk the audience will: Understand how GPUs differ from CPUs Understand the limits of GPU acceleration Have

More information

Energy-efficient acceleration of task dependency trees on CPU-GPU hybrids

Energy-efficient acceleration of task dependency trees on CPU-GPU hybrids Energy-efficient acceleration of task dependency trees on CPU-GPU hybrids Mark Silberstein - Technion Naoya Maruyama Tokyo Institute of Technology Mark Silberstein, Technion 1 The case for heterogeneous

More information

Parallelization of the QR Decomposition with Column Pivoting Using Column Cyclic Distribution on Multicore and GPU Processors

Parallelization of the QR Decomposition with Column Pivoting Using Column Cyclic Distribution on Multicore and GPU Processors Parallelization of the QR Decomposition with Column Pivoting Using Column Cyclic Distribution on Multicore and GPU Processors Andrés Tomás 1, Zhaojun Bai 1, and Vicente Hernández 2 1 Department of Computer

More information

Flux Vector Splitting Methods for the Euler Equations on 3D Unstructured Meshes for CPU/GPU Clusters

Flux Vector Splitting Methods for the Euler Equations on 3D Unstructured Meshes for CPU/GPU Clusters Flux Vector Splitting Methods for the Euler Equations on 3D Unstructured Meshes for CPU/GPU Clusters Manfred Liebmann Technische Universität München Chair of Optimal Control Center for Mathematical Sciences,

More information

Scientific Computations Using Graphics Processors

Scientific Computations Using Graphics Processors Scientific Computations Using Graphics Processors Blair Perot Ali Khajeh-Saeed Tim McGuiness History Kevin Bowers, X Division Los Alamos Lab (2003) Lots of Memory Uses Memory Banks Cheap (commodity) Relativistic

More information

Automatic Tuning of the High Performance Linpack Benchmark

Automatic Tuning of the High Performance Linpack Benchmark Automatic Tuning of the High Performance Linpack Benchmark Ruowei Chen Supervisor: Dr. Peter Strazdins The Australian National University What is the HPL Benchmark? World s Top 500 Supercomputers http://www.top500.org

More information

Optimization solutions for the segmented sum algorithmic function

Optimization solutions for the segmented sum algorithmic function Optimization solutions for the segmented sum algorithmic function ALEXANDRU PÎRJAN Department of Informatics, Statistics and Mathematics Romanian-American University 1B, Expozitiei Blvd., district 1, code

More information

GPU-Accelerated Parallel Sparse LU Factorization Method for Fast Circuit Analysis

GPU-Accelerated Parallel Sparse LU Factorization Method for Fast Circuit Analysis GPU-Accelerated Parallel Sparse LU Factorization Method for Fast Circuit Analysis Abstract: Lower upper (LU) factorization for sparse matrices is the most important computing step for circuit simulation

More information

Di Zhao Ohio State University MVAPICH User Group (MUG) Meeting, August , Columbus Ohio

Di Zhao Ohio State University MVAPICH User Group (MUG) Meeting, August , Columbus Ohio Di Zhao zhao.1029@osu.edu Ohio State University MVAPICH User Group (MUG) Meeting, August 26-27 2013, Columbus Ohio Nvidia Kepler K20X Intel Xeon Phi 7120 Launch Date November 2012 Q2 2013 Processor Per-processor

More information

Accelerating Financial Applications on the GPU

Accelerating Financial Applications on the GPU Accelerating Financial Applications on the GPU Scott Grauer-Gray Robert Searles William Killian John Cavazos Department of Computer and Information Science University of Delaware Sixth Workshop on General

More information

A Framework for Modeling GPUs Power Consumption

A Framework for Modeling GPUs Power Consumption A Framework for Modeling GPUs Power Consumption Sohan Lal, Jan Lucas, Michael Andersch, Mauricio Alvarez-Mesa, Ben Juurlink Embedded Systems Architecture Technische Universität Berlin Berlin, Germany January

More information

Technology for a better society. hetcomp.com

Technology for a better society. hetcomp.com Technology for a better society hetcomp.com 1 J. Seland, C. Dyken, T. R. Hagen, A. R. Brodtkorb, J. Hjelmervik,E Bjønnes GPU Computing USIT Course Week 16th November 2011 hetcomp.com 2 9:30 10:15 Introduction

More information

Massively Parallel Computing with CUDA. Carlos Alberto Martínez Angeles Cinvestav-IPN

Massively Parallel Computing with CUDA. Carlos Alberto Martínez Angeles Cinvestav-IPN Massively Parallel Computing with CUDA Carlos Alberto Martínez Angeles Cinvestav-IPN What is a GPU? A graphics processing unit (GPU) The term GPU was popularized by Nvidia in 1999 marketed the GeForce

More information

Accelerating HPL on Heterogeneous GPU Clusters

Accelerating HPL on Heterogeneous GPU Clusters Accelerating HPL on Heterogeneous GPU Clusters Presentation at GTC 2014 by Dhabaleswar K. (DK) Panda The Ohio State University E-mail: panda@cse.ohio-state.edu http://www.cse.ohio-state.edu/~panda Outline

More information

Accelerating Leukocyte Tracking Using CUDA: A Case Study in Leveraging Manycore Coprocessors

Accelerating Leukocyte Tracking Using CUDA: A Case Study in Leveraging Manycore Coprocessors Accelerating Leukocyte Tracking Using CUDA: A Case Study in Leveraging Manycore Coprocessors Michael Boyer, David Tarjan, Scott T. Acton, and Kevin Skadron University of Virginia IPDPS 2009 Outline Leukocyte

More information

POST-SIEVING ON GPUs

POST-SIEVING ON GPUs POST-SIEVING ON GPUs Andrea Miele 1, Joppe W Bos 2, Thorsten Kleinjung 1, Arjen K Lenstra 1 1 LACAL, EPFL, Lausanne, Switzerland 2 NXP Semiconductors, Leuven, Belgium 1/18 NUMBER FIELD SIEVE (NFS) Asymptotically

More information

Recent Advances in Heterogeneous Computing using Charm++

Recent Advances in Heterogeneous Computing using Charm++ Recent Advances in Heterogeneous Computing using Charm++ Jaemin Choi, Michael Robson Parallel Programming Laboratory University of Illinois Urbana-Champaign April 12, 2018 1 / 24 Heterogeneous Computing

More information

Accelerating Implicit LS-DYNA with GPU

Accelerating Implicit LS-DYNA with GPU Accelerating Implicit LS-DYNA with GPU Yih-Yih Lin Hewlett-Packard Company Abstract A major hindrance to the widespread use of Implicit LS-DYNA is its high compute cost. This paper will show modern GPU,

More information

GPU & High Performance Computing (by NVIDIA) CUDA. Compute Unified Device Architecture Florian Schornbaum

GPU & High Performance Computing (by NVIDIA) CUDA. Compute Unified Device Architecture Florian Schornbaum GPU & High Performance Computing (by NVIDIA) CUDA Compute Unified Device Architecture 29.02.2008 Florian Schornbaum GPU Computing Performance In the last few years the GPU has evolved into an absolute

More information