O.I. Streltsova, D.V. Podgainy, M.V. Bashashin, M.I.Zuev

Size: px
Start display at page:

Download "O.I. Streltsova, D.V. Podgainy, M.V. Bashashin, M.I.Zuev"

Transcription

1 High Performance Computing Technologies Lecture, Practical training 9 Parallel Computing with MPI: parallel algorithm for linear algebra O.I. Streltsova, D.V. Podgainy, M.V. Bashashin, M.I.Zuev Heterogeneous Computations Team, HybriLIT Laboratory of Information Technologies Joint Institute for Nuclear Research Dubna University 12 April 2018 Heterogeneous Computation Team, HybriLIT

2 Serial computing Traditionally, software has been written for serial computation: To be run on a single computer having a single Central Processing Unit (CPU); A problem is broken into a discrete series of instructions. Instructions are executed one after another. Only one instruction may execute at any moment in time.

3 Parallel computing Parallel computing is the simultaneous use of multiple compute resources to solve a computational problem: To be run using multiple CPUs. A problem is broken into discrete parts that can be solved concurrently. Each part is further broken down to a series of instructions. Instructions from each part execute simultaneously on different CPUs.

4 Types of parallel machines Distributed memory each processor has its own memory address space. Examples: clusters, Blue Gene/L Shared memory single address space for all processors. Examples: IBM-series, multi-core PC Uniform Memory Access CPU0 CPU1 CPU2 CPU0 Mem0 Mem1 CPU1 CPU2 Memory Mem2 Non-Uniform Memory Access MPI is a library standard for programming distributed memory system CPU CPU Memory CPU CPU Memory CPU CPU Memory CPU CPU Memory

5 MPI. Main conception MPI: Message Passing Interface Data transfer operation MPI is based on data transfer operations. Among MPI functions there are: point-to-point operations between two processes, collective communication actions for simultaneous interaction of several processes. 1. Grama, Gupta, Karypis, Kumar. Introduction to Parallel Computing, Second Edition Snir, Otto, Huss-Lederman, Walker, Dongarra. MPI: The Complete Reference, Volume 1, The MPI Core, Second edition. 1998

6 MPI: Concepts Processes Group of processes Communicator MPI COMM WORLD name of the default MPI communicator: MPI_Comm_size (MPI_COMM_WORLD, &size); MPI_Comm_rank(MPI_COMM_WORLD, &rank); MPI-Communicator: MPI_COMM_WORLD Process 0 Process 2 Process 1 Process size-1 Each process in a communicator is identified by its rank

7 The 6 most important MPI commands MPI_Init() initiate a MPI computation MPI_Comm_size() how many processes participate in a given MPI communicator? MPI_Comm_rank() which one am I? MPI_Send() send a message MPI_Recv() receive a message (A number between 0 and size-1) MPI_Finalize() terminate the MPI computation and clean up

8 MPI. Basic Datatypes At message passing operation execution, it is necessary to designate the type of passed data on order to denote the passed or received data. MPI contains a big set of base data types that coincide with datatypes in C/C++ and Fortran programming languages. MPI MPI_CHAR MPI_DOUBLE MPI_FLOAT MPI_INT C/C++ signed char double float int All using MPI datatypes you can see in file mpi.h (mpif.h) MPI includes possibilities for creation of new derived data types for more detailed and short description of sent messages.

9 Basic Needs in parallel programming In order to do parallel programming, we need basic functionality: Start Processes Send Messages Receive Messages Synchronize processes

10 Language notation Fortran include "mpif.h" call MPI_FUNCTION(parameter,..., ierr) Compilation $ mpif77 prog_name.f o prog_exec C/C++ #include <mpi.h> MPI_Function(parameter,...); Compilation $ mpicc/mpic++ prog_name.c o prog_exec

11 Example MPI-prorgam mpi_hello.c #include <stdio.h> #include <mpi.h> int main(int argc, char **argv){ int size, rank; MPI_Init(&argc,&argv); Initiate an MPI computation MPI_Comm_size (MPI_COMM_WORLD, &size); How many process? MPI_Comm_rank(MPI_COMM_WORLD, &rank); The rank value is between 0 and size-1 printf("hello world! I am process number: %d form %s\n", rank, size); MPI_Finalize(); Terminate the MPI computation and clean up return 0;

12 Compilation, execution on HybriLIT Add module $ module add openmpi/1.8.1 Compilation $ mpicc -std=c99 mpi_hello.c -o exec_mpi script_mpi #!/bin/sh #SBATCH -p tut #SBATCH -n 8 #SBATCH -t 60 mpiexec./exec_mpi Run in batch mode $ sbatch script_mpi

13 Output, parallel execution Output Numprocs is 6; Hello from 0 Numprocs is 6; Hello from 1 Numprocs is 6; Hello from 5 Numprocs is 6; Hello from 3 Numprocs is 6; Hello from 4 Numprocs is 6; Hello from 2

14 MPI point-to-point communications int MPI_Send(void *buf, // initial address of send buffer int count, // number of elements in send buffer MPI_Datatype datatype, // datatype of each send elements int dest, // rank of destination int tag, // message tag MPI_Comm comm); // communicator

15 MPI point-to-point communications int MPI_Recv(void *buf, // initial address of receive buffer int count, // number of elements in send buffer MPI_Datatype datatype, // datatype of each send elements int source, // rank of source int tag, // message tag MPI_Comm comm, // communicator MPI_Status *status); // status object

16 MPI point-to-point communications. Example mpi_send-recv.c #include <mpi.h> #include <stdio.h> #include <string.h> int main(int argc, char **argv){ int size, rank, tag,rc, i; char mess[30]; MPI_Status status; rc= MPI_Init(&argc,&argv); if (rc!= MPI_SUCCESS) { printf("error starting MPI program. Terminate.\n"); MPI_Abort(MPI_COMM_WORLD, rc);

17 MPI point-to-point communications. Example MPI_Comm_size (MPI_COMM_WORLD, &size); MPI_Comm_rank(MPI_COMM_WORLD, &rank); tag= 8; if (rank==0){ strcpy(mess, "Hello from process 0!"); for (i=1; i <size; i++) MPI_Send(mess, 21, MPI_CHAR, i, tag, MPI_COMM_WORLD); else{ MPI_Recv(mess, 21, MPI_CHAR, 0, tag, MPI_COMM_WORLD, &status); printf("process number: %d message= %s \n", rank, mess); MPI_Finalize(); return 0;

18 Jacobi method in solution of SLAE A x f, A R M M System of linear algebraic equations M a a, i 1, M ii j 1, j i ij Condition of diagonally domination of matrix M k fi aij x j k 1 j 1, j i xi i M k aii M n 1 n n 1 n i i i 1, 1, ; 0 2 x x x x Computation the next approximation k number of iteration Criterion of ending iteration process S.A. Lupin, M.A. Posypkin. Parallel programming. M.: ID FORUM, INFRA-M, p.

19 Used functions void init (double* a, double* f, double* x){ for(int i= 0; i<m; i++){ f[i]= 3.0*M - 1.0; for(int j= 0; j < M; j++){ if(i == j) a[i*m+i]= 2.0*M; else a[i*m+j]= 1.0; for(int i= 0; i<m; i++) x[i]= 0.0; x[0]= 1.0; Initialization of matrix A x f, A R M M

20 Used functions Computational of criterion of ending iteration process double evaldiff (double* u, double* v, int m){ double d= 0.0; for(int i= 0; i<m; i++){ double b; b= v[i]-u[i]; d+= b*b; return sqrt(d); M n 1 n n 1 n i i i 1 2 x x x x

21 Used functions double matvec (double* a, double* f, double* xold, double* x, int m, int n){ for(int i= 1; i<m; i++){ double sum= 0.0; for(int j= 0; j<n; j++){ if(i!=j) sum+= a[i*m+j]*xold[j]; x[i]= (f[i]-sum)/a[i*m+i]; Computation function of the next approximation M k fi aij x j k 1 j 1, j i xi i M k aii, 1, ; 0

22 Serial code jacobi_01_serial.c #include <stdio.h> #include <stdlib.h> #include <string.h> #include <time.h> #include <math.h> #define MAX_ITERS 200 // The maximum number of iterations #define M // The dimension of the matrix #define EPS 1e-5 // The accuracy of calculations

23 Serial code int main(int argc, char* argv[]){ // I count of iterations int I= 0; // xold previous approximation x(k) // x next approximation x(k+1) // xexact exact solution // diff norm of the difference // t program runtime double *xold, *x, *xexact, *a, *f, diff, t;

24 Serial code // Allocate memory for the matrix and vectors xold= (double*)malloc(m*sizeof(double)); x= (double*)malloc(m*sizeof(double)); xexact= (double*)malloc(m*sizeof(double)); a= (double*)malloc(m*m*sizeof(double)); f= (double*)malloc(m*sizeof(double)); init(a, f, x);

25 Serial code for(int i= 0; i<m; i++) xexact[i]= 1.0; printf(" Size of the matrix= %d, Accuracy= %5.3e\n", M, EPS); t= time(null); // Main loop do{ matvec(a, f, xold, x, M, M); diff= evaldiff(xold, x, M); I++; printf("diff= %8.5f\n", diff);

26 Serial code // Copy the next approximation to the array of the previous memcpy(xold, x, M*sizeof(double)); while((diff >= EPS) && (I <= MAX_ITERS)); // Check for reaching a maximum number of iterations if(i>=max_iters){ printf("reached a maximum number of iterations\n");

27 Serial code double err1= 0.0; double err2= 0.0; for(int i= 0; i<m; i++){ double tmp= fabs(x[i]-xexact[i]); err2+= (x[i]-xexact[i])*(x[i]-xexact[i]); if(tmp > err1) err1= tmp; double err2sqrt= sqrt(err2); // Calculating program runtime t= time(null)-t;

28 Serial code printf("error= %12.9f\n", err1); printf("norma= %12.9f\n", err2sqrt); // Output the number of iterations and program runtime printf("%d iterations consumed %lf sec\n", I, t); // Free memory free(xold); free(x); free(xexact); free(a); free(f); return 0;

29 MPI collective communication int MPI_Bcast( void *buffer, int count, MPI_Datatype datatype, int root, MPI_Comm comm );

30 MPI collective communication int MPI_Scatter( const void *sendbuf, int sendcount, MPI_Datatype sendtype, void *recvbuf, int recvcount, MPI_Datatype recvtype, int root, MPI_Comm comm );

31 MPI collective communication int MPI_Gather( const void *sendbuf, int sendcount, MPI_Datatype sendtype, void *recvbuf, int recvcount, MPI_Datatype recvtype, int root, MPI_Comm comm );

32 Update used function Computation function of the next approximation double matvec (double* a, double* f, double* xold, double* x, int rk, int sp, int m, int n){ for(int i= 1; i<m; i++){ double sum= 0.0; for(int j= 0; j<n; j++){ if(j!= i+rk*sp) sum+= a[i*m+j]*xold[j]; x[i]= (f[i]-sum)/a[i*m+i+rk*sp];

33 MPI code. Version 1 jacobi_02_mpi.c #include... #include <mpi.h> // Include header for support MPI library #define... int main(int argc, char* argv[]){ // size total number of processes // rank rank of the process // chunk the number of rows processed by each process int I= 0, size, rank, chunk; // aloc array to store the local part of the vector a // floc array to store the local part of the vector f // xloc array to store the local part of the vector x double *a, *aloc, *f, *floc, *xold, *x, *xloc, diff, t;

34 MPI code. Version 1 // Initialization of MPI MPI_Init(&argc, &argv); MPI_Comm_size(MPI_COMM_WORLD, &size); MPI_Comm_rank(MPI_COMM_WORLD, &rank); if(rank==0){ // Work only root process printf(" Simple MPI realization\n"); printf(" Number of processes= %d\n Size of the matrix= %d, Accuracy= %5.3e\n", size, M, EPS); chunk= M/size; a= (double*)malloc(m*m*sizeof(double));

35 MPI code. Version 1 // Broadcasting value of chunk to all processes MPI_Bcast(&chunk, 1, MPI_INT, 0, MPI_COMM_WORLD); // Allocation memory for arrays and vectors aloc= (double*)malloc(chunk*m*sizeof(double)); f= (double*)malloc(m*sizeof(double)); floc= (double*)malloc(chunk*sizeof(double)); xold= (double*)malloc(m*sizeof(double)); x= (double*)malloc(m*sizeof(double)); xloc= (double*)malloc(chunk*sizeof(double));

36 MPI code. Version 1 if(rank==0){ init(a, f, xold); t= MPI_Wtime(); // Send array A and the vector f to processes MPI_Scatter(a, M*chunk, MPI_DOUBLE, aloc, M*chunk, MPI_DOUBLE, 0, MPI_COMM_WORLD); MPI_Scatter(f, chunk, MPI_DOUBLE, floc, chunk, MPI_DOUBLE, 0, MPI_COMM_WORLD);

37 Parallel code

38 MPI code. Version 1 // Main loop do{ // Broadcasting value of the previous approximation to all processes MPI_Bcast(xold, M, MPI_DOUBLE, 0, MPI_COMM_WORLD); matvec(aloc, floc, xold, xloc, rank, chunk, chunk, M); // Gather of vector x at root process MPI_Gather(xloc, chunk, MPI_DOUBLE, x, chunk, MPI_DOUBLE, 0, MPI_COMM_WORLD);

39 MPI code. Version 1 if(rank==0){ diff= evaldiff(x, xold, M); printf("diff= %10.7f\n", diff); memcpy(xold, x, M*sizeof(double)); MPI_Bcast(&diff, 1, MPI_DOUBLE, 0, MPI_COMM_WORLD); I++; while((diff >= EPS) && (I <= MAX_ITERS));

40 Parallel code

41 MPI code. Version 1 // Output the number of iterations and program runtime if(rank==0){ t= MPI_Wtime()-t; printf("%d iterations consumed %lf sec\n", I, t); // Free memory if(rank==0) free(a); free(aloc); free(f); free(floc); free(xold); free(x); free(xloc); // Finalize of all MPI processes MPI_Finalize(); return 0;

42 MPI code. Version 1. Optimization void init(double* f, double* x){ // f right hand side vector // x vector of initial approximation to the solution for(int i= 0; i<m; i++){ f[i]= 3.0*M - 1.0; for(int i= 0; i<m; i++) x[i]= 0.0; x[0]= 1.0;

43 MPI code. Version 1. Optimization jacobi_02_mpi_alloc.c // Init of matrix aloc on each process for(int i= 0; i<chunk; i++){ for(int j= 0; j<m; j++){ if(j == i+rank*chunk) aloc[i*m+j]= 2.0*M; else aloc[i*m+j]= 1.0;

44 MPI collective communication int MPI_Scatterv( const void *sendbuf, const int *sendcounts, const int *displs, MPI_Datatype sendtype, void *recvbuf, int recvcount, MPI_Datatype recvtype, int root, MPI_Comm comm );

45 MPI collective communication int MPI_Gatherv( const void *sendbuf, int sendcount, MPI_Datatype sendtype, void *recvbuf, const int *recvcounts, const int *displs, MPI_Datatype recvtype, int root, MPI_Comm comm );

46 MPI code. Version 2 jacobi_03_scatterv_gatherv.c #include... #define... int main(int argc, char* argv[]){ // chunks an array, each element of which contains the number of elements in the respective data block // disps an array containing the offset value measured in the sent items int I= 0, size, rank, chunk, *chunks, *disps; double *a, *aloc, *f, *floc, *xold, *x, *xloc, *xexact, *xtemp, diff, t;

47 MPI code. Version 2 // Initialization of MPI MPI_Init(&argc, &argv); MPI_Comm_size(MPI_COMM_WORLD, &size); MPI_Comm_rank(MPI_COMM_WORLD, &rank); if(rank==0){ printf(" MPI realization with Scatterv() and Gatherv()\n"); printf(" Number of processes= %d\n Size of the matrix= %d, Accuracy= %5.3e\n", size, M, EPS); chunk= M/size; a= (double*)malloc(m*m*sizeof(double)); // Broadcasting value of "chunk" to all processes MPI_Bcast(&chunk, 1, MPI_INT, 0, MPI_COMM_WORLD);

48 MPI code. Version 2 // Allocation memory for arrays and vectors chunks= (int*)malloc(size*sizeof(int)); disps= (int*)malloc(size*sizeof(int)); f= (double*)malloc(m*sizeof(double)); x= (double*)malloc(m*sizeof(double)); xold= (double*)malloc(m*sizeof(double)); if(rank==0){ init(a, f, xold); t= MPI_Wtime();

49 MPI code. Version 2 for(int i= 0; i<size; i++){ disps[i]= i*chunk*m; if(i==(size-1)) else chunks[i]= (M-(size-1)*chunk)*M; chunks[i]= chunk*m; aloc= (double*)malloc(chunks[rank]*sizeof(double)); MPI_Scatterv(a, chunks, disps, MPI_DOUBLE, aloc, chunks[rank], MPI_DOUBLE, 0, MPI_COMM_WORLD);

50 MPI code. Version 2 for(int i= 0; i<size; i++){ disps[i]= i*chunk; if(i==(size-1)) else chunks[i]= M-(size-1)*chunk; chunks[i]= chunk; floc= (double*)malloc(chunks[rank]*sizeof(double)); MPI_Scatterv(f, chunks, disps, MPI_DOUBLE, floc, chunks[rank], MPI_DOUBLE, 0, MPI_COMM_WORLD);

51 MPI code. Version 2 xloc= (double*)malloc(chunks[rank]*sizeof(double)); // Main loop do{ MPI_Bcast(xold, M, MPI_DOUBLE, 0, MPI_COMM_WORLD); matvec(aloc, floc, xold, xloc, rank, chunk, chunks[rank], M); MPI_Gatherv(xloc, chunks[rank], MPI_DOUBLE, x, chunks, disps, MPI_DOUBLE, 0, MPI_COMM_WORLD); if(rank==0){ diff= evaldiff(x, xold, M); printf("diff= %10.7f\n", diff); memcpy(xold, x, M*sizeof(double));

52 MPI code. Version 2 MPI_Bcast(&diff, 1, MPI_DOUBLE, 0, MPI_COMM_WORLD); I++; while((diff >= EPS) && (I <= MAX_ITERS)); // Output the number of iterations and program runtime if(rank==0){ t= MPI_Wtime()-t; printf("%d iterations consumed %lf sec\n", I, t); // Free memory if(rank==0) free(a); free(chunks); free(disps); free(f); free(x); free(xold); free(aloc); free(floc); free(xloc); // Finalize of all MPI processes MPI_Finalize(); return 0;

53 MPI collective communication int MPI_Allgatherv( const void *sendbuf, int sendcount, MPI_Datatype sendtype, void *recvbuf, const int *recvcounts, const int *displs, MPI_Datatype recvtype, MPI_Comm comm)

54 do{ MPI code. Version 3 The main loop will look like the following. The remaining part of the code will not change. jacobi_04_allgatherv.c matvec(aloc, floc, xold, xloc, rank, chunk, chunks[rank], M); MPI_Allgatherv(xloc, chunks[rank], MPI_DOUBLE, x, chunks, disps, MPI_DOUBLE, MPI_COMM_WORLD); diff= evaldiff(x, xold, M); memcpy(xold, x, M*sizeof(double)); I++; while((diff >= EPS) && (I <= MAX_ITERS));

55 Practical Tasks - Carry out computations at different number of MPI-processes and plot graphs on speedup and efficiency. - Perform broadcast transfer from each processor by means of peer-to-peer messaging

56

MPI. (message passing, MIMD)

MPI. (message passing, MIMD) MPI (message passing, MIMD) What is MPI? a message-passing library specification extension of C/C++ (and Fortran) message passing for distributed memory parallel programming Features of MPI Point-to-point

More information

MPI Collective communication

MPI Collective communication MPI Collective communication CPS343 Parallel and High Performance Computing Spring 2018 CPS343 (Parallel and HPC) MPI Collective communication Spring 2018 1 / 43 Outline 1 MPI Collective communication

More information

CS 470 Spring Mike Lam, Professor. Distributed Programming & MPI

CS 470 Spring Mike Lam, Professor. Distributed Programming & MPI CS 470 Spring 2017 Mike Lam, Professor Distributed Programming & MPI MPI paradigm Single program, multiple data (SPMD) One program, multiple processes (ranks) Processes communicate via messages An MPI

More information

HPC Parallel Programing Multi-node Computation with MPI - I

HPC Parallel Programing Multi-node Computation with MPI - I HPC Parallel Programing Multi-node Computation with MPI - I Parallelization and Optimization Group TATA Consultancy Services, Sahyadri Park Pune, India TCS all rights reserved April 29, 2013 Copyright

More information

Message Passing Interface

Message Passing Interface MPSoC Architectures MPI Alberto Bosio, Associate Professor UM Microelectronic Departement bosio@lirmm.fr Message Passing Interface API for distributed-memory programming parallel code that runs across

More information

CS 470 Spring Mike Lam, Professor. Distributed Programming & MPI

CS 470 Spring Mike Lam, Professor. Distributed Programming & MPI CS 470 Spring 2018 Mike Lam, Professor Distributed Programming & MPI MPI paradigm Single program, multiple data (SPMD) One program, multiple processes (ranks) Processes communicate via messages An MPI

More information

Outline. Communication modes MPI Message Passing Interface Standard

Outline. Communication modes MPI Message Passing Interface Standard MPI THOAI NAM Outline Communication modes MPI Message Passing Interface Standard TERMs (1) Blocking If return from the procedure indicates the user is allowed to reuse resources specified in the call Non-blocking

More information

Introduction to MPI. Ricardo Fonseca. https://sites.google.com/view/rafonseca2017/

Introduction to MPI. Ricardo Fonseca. https://sites.google.com/view/rafonseca2017/ Introduction to MPI Ricardo Fonseca https://sites.google.com/view/rafonseca2017/ Outline Distributed Memory Programming (MPI) Message Passing Model Initializing and terminating programs Point to point

More information

Introduction to parallel computing concepts and technics

Introduction to parallel computing concepts and technics Introduction to parallel computing concepts and technics Paschalis Korosoglou (support@grid.auth.gr) User and Application Support Unit Scientific Computing Center @ AUTH Overview of Parallel computing

More information

Introduction to the Message Passing Interface (MPI)

Introduction to the Message Passing Interface (MPI) Introduction to the Message Passing Interface (MPI) CPS343 Parallel and High Performance Computing Spring 2018 CPS343 (Parallel and HPC) Introduction to the Message Passing Interface (MPI) Spring 2018

More information

Introduction to MPI. HY555 Parallel Systems and Grids Fall 2003

Introduction to MPI. HY555 Parallel Systems and Grids Fall 2003 Introduction to MPI HY555 Parallel Systems and Grids Fall 2003 Outline MPI layout Sending and receiving messages Collective communication Datatypes An example Compiling and running Typical layout of an

More information

Outline. Communication modes MPI Message Passing Interface Standard. Khoa Coâng Ngheä Thoâng Tin Ñaïi Hoïc Baùch Khoa Tp.HCM

Outline. Communication modes MPI Message Passing Interface Standard. Khoa Coâng Ngheä Thoâng Tin Ñaïi Hoïc Baùch Khoa Tp.HCM THOAI NAM Outline Communication modes MPI Message Passing Interface Standard TERMs (1) Blocking If return from the procedure indicates the user is allowed to reuse resources specified in the call Non-blocking

More information

CS 426. Building and Running a Parallel Application

CS 426. Building and Running a Parallel Application CS 426 Building and Running a Parallel Application 1 Task/Channel Model Design Efficient Parallel Programs (or Algorithms) Mainly for distributed memory systems (e.g. Clusters) Break Parallel Computations

More information

Recap of Parallelism & MPI

Recap of Parallelism & MPI Recap of Parallelism & MPI Chris Brady Heather Ratcliffe The Angry Penguin, used under creative commons licence from Swantje Hess and Jannis Pohlmann. Warwick RSE 13/12/2017 Parallel programming Break

More information

Distributed Memory Programming with MPI

Distributed Memory Programming with MPI Distributed Memory Programming with MPI Moreno Marzolla Dip. di Informatica Scienza e Ingegneria (DISI) Università di Bologna moreno.marzolla@unibo.it Algoritmi Avanzati--modulo 2 2 Credits Peter Pacheco,

More information

Lesson 1. MPI runs on distributed memory systems, shared memory systems, or hybrid systems.

Lesson 1. MPI runs on distributed memory systems, shared memory systems, or hybrid systems. The goals of this lesson are: understanding the MPI programming model managing the MPI environment handling errors point-to-point communication 1. The MPI Environment Lesson 1 MPI (Message Passing Interface)

More information

Distributed Memory Programming with Message-Passing

Distributed Memory Programming with Message-Passing Distributed Memory Programming with Message-Passing Pacheco s book Chapter 3 T. Yang, CS240A Part of slides from the text book and B. Gropp Outline An overview of MPI programming Six MPI functions and

More information

Message Passing Interface

Message Passing Interface Message Passing Interface DPHPC15 TA: Salvatore Di Girolamo DSM (Distributed Shared Memory) Message Passing MPI (Message Passing Interface) A message passing specification implemented

More information

Introduction to MPI. Ekpe Okorafor. School of Parallel Programming & Parallel Architecture for HPC ICTP October, 2014

Introduction to MPI. Ekpe Okorafor. School of Parallel Programming & Parallel Architecture for HPC ICTP October, 2014 Introduction to MPI Ekpe Okorafor School of Parallel Programming & Parallel Architecture for HPC ICTP October, 2014 Topics Introduction MPI Model and Basic Calls MPI Communication Summary 2 Topics Introduction

More information

Message Passing Interface. most of the slides taken from Hanjun Kim

Message Passing Interface. most of the slides taken from Hanjun Kim Message Passing Interface most of the slides taken from Hanjun Kim Message Passing Pros Scalable, Flexible Cons Someone says it s more difficult than DSM MPI (Message Passing Interface) A standard message

More information

CSE 613: Parallel Programming. Lecture 21 ( The Message Passing Interface )

CSE 613: Parallel Programming. Lecture 21 ( The Message Passing Interface ) CSE 613: Parallel Programming Lecture 21 ( The Message Passing Interface ) Jesmin Jahan Tithi Department of Computer Science SUNY Stony Brook Fall 2013 ( Slides from Rezaul A. Chowdhury ) Principles of

More information

In the simplest sense, parallel computing is the simultaneous use of multiple computing resources to solve a problem.

In the simplest sense, parallel computing is the simultaneous use of multiple computing resources to solve a problem. 1. Introduction to Parallel Processing In the simplest sense, parallel computing is the simultaneous use of multiple computing resources to solve a problem. a) Types of machines and computation. A conventional

More information

Parallel programming MPI

Parallel programming MPI Parallel programming MPI Distributed memory Each unit has its own memory space If a unit needs data in some other memory space, explicit communication (often through network) is required Point-to-point

More information

Practical Course Scientific Computing and Visualization

Practical Course Scientific Computing and Visualization July 5, 2006 Page 1 of 21 1. Parallelization Architecture our target architecture: MIMD distributed address space machines program1 data1 program2 data2 program program3 data data3.. program(data) program1(data1)

More information

Introduction to Parallel and Distributed Systems - INZ0277Wcl 5 ECTS. Teacher: Jan Kwiatkowski, Office 201/15, D-2

Introduction to Parallel and Distributed Systems - INZ0277Wcl 5 ECTS. Teacher: Jan Kwiatkowski, Office 201/15, D-2 Introduction to Parallel and Distributed Systems - INZ0277Wcl 5 ECTS Teacher: Jan Kwiatkowski, Office 201/15, D-2 COMMUNICATION For questions, email to jan.kwiatkowski@pwr.edu.pl with 'Subject=your name.

More information

CS 179: GPU Programming. Lecture 14: Inter-process Communication

CS 179: GPU Programming. Lecture 14: Inter-process Communication CS 179: GPU Programming Lecture 14: Inter-process Communication The Problem What if we want to use GPUs across a distributed system? GPU cluster, CSIRO Distributed System A collection of computers Each

More information

The Message Passing Model

The Message Passing Model Introduction to MPI The Message Passing Model Applications that do not share a global address space need a Message Passing Framework. An application passes messages among processes in order to perform

More information

Distributed Systems + Middleware Advanced Message Passing with MPI

Distributed Systems + Middleware Advanced Message Passing with MPI Distributed Systems + Middleware Advanced Message Passing with MPI Gianpaolo Cugola Dipartimento di Elettronica e Informazione Politecnico, Italy cugola@elet.polimi.it http://home.dei.polimi.it/cugola

More information

MPI: Parallel Programming for Extreme Machines. Si Hammond, High Performance Systems Group

MPI: Parallel Programming for Extreme Machines. Si Hammond, High Performance Systems Group MPI: Parallel Programming for Extreme Machines Si Hammond, High Performance Systems Group Quick Introduction Si Hammond, (sdh@dcs.warwick.ac.uk) WPRF/PhD Research student, High Performance Systems Group,

More information

MPI Message Passing Interface

MPI Message Passing Interface MPI Message Passing Interface Portable Parallel Programs Parallel Computing A problem is broken down into tasks, performed by separate workers or processes Processes interact by exchanging information

More information

CS4961 Parallel Programming. Lecture 18: Introduction to Message Passing 11/3/10. Final Project Purpose: Mary Hall November 2, 2010.

CS4961 Parallel Programming. Lecture 18: Introduction to Message Passing 11/3/10. Final Project Purpose: Mary Hall November 2, 2010. Parallel Programming Lecture 18: Introduction to Message Passing Mary Hall November 2, 2010 Final Project Purpose: - A chance to dig in deeper into a parallel programming model and explore concepts. -

More information

Introduction to MPI. Jerome Vienne Texas Advanced Computing Center January 10 th,

Introduction to MPI. Jerome Vienne Texas Advanced Computing Center January 10 th, Introduction to MPI Jerome Vienne Texas Advanced Computing Center January 10 th, 2013 Email: viennej@tacc.utexas.edu 1 Course Objectives & Assumptions Objectives Teach basics of MPI-Programming Share information

More information

Holland Computing Center Kickstart MPI Intro

Holland Computing Center Kickstart MPI Intro Holland Computing Center Kickstart 2016 MPI Intro Message Passing Interface (MPI) MPI is a specification for message passing library that is standardized by MPI Forum Multiple vendor-specific implementations:

More information

PCAP Assignment I. 1. A. Why is there a large performance gap between many-core GPUs and generalpurpose multicore CPUs. Discuss in detail.

PCAP Assignment I. 1. A. Why is there a large performance gap between many-core GPUs and generalpurpose multicore CPUs. Discuss in detail. PCAP Assignment I 1. A. Why is there a large performance gap between many-core GPUs and generalpurpose multicore CPUs. Discuss in detail. The multicore CPUs are designed to maximize the execution speed

More information

MPI MPI. Linux. Linux. Message Passing Interface. Message Passing Interface. August 14, August 14, 2007 MPICH. MPI MPI Send Recv MPI

MPI MPI. Linux. Linux. Message Passing Interface. Message Passing Interface. August 14, August 14, 2007 MPICH. MPI MPI Send Recv MPI Linux MPI Linux MPI Message Passing Interface Linux MPI Linux MPI Message Passing Interface MPI MPICH MPI Department of Science and Engineering Computing School of Mathematics School Peking University

More information

Parallel Computing and the MPI environment

Parallel Computing and the MPI environment Parallel Computing and the MPI environment Claudio Chiaruttini Dipartimento di Matematica e Informatica Centro Interdipartimentale per le Scienze Computazionali (CISC) Università di Trieste http://www.dmi.units.it/~chiarutt/didattica/parallela

More information

Introduction to MPI. Ritu Arora Texas Advanced Computing Center June 17,

Introduction to MPI. Ritu Arora Texas Advanced Computing Center June 17, Introduction to MPI Ritu Arora Texas Advanced Computing Center June 17, 2014 Email: rauta@tacc.utexas.edu 1 Course Objectives & Assumptions Objectives Teach basics of MPI-Programming Share information

More information

Message Passing Interface

Message Passing Interface Message Passing Interface by Kuan Lu 03.07.2012 Scientific researcher at Georg-August-Universität Göttingen and Gesellschaft für wissenschaftliche Datenverarbeitung mbh Göttingen Am Faßberg, 37077 Göttingen,

More information

An Introduction to MPI

An Introduction to MPI An Introduction to MPI Parallel Programming with the Message Passing Interface William Gropp Ewing Lusk Argonne National Laboratory 1 Outline Background The message-passing model Origins of MPI and current

More information

Parallel Programming with MPI: Day 1

Parallel Programming with MPI: Day 1 Parallel Programming with MPI: Day 1 Science & Technology Support High Performance Computing Ohio Supercomputer Center 1224 Kinnear Road Columbus, OH 43212-1163 1 Table of Contents Brief History of MPI

More information

Introduction to MPI. SHARCNET MPI Lecture Series: Part I of II. Paul Preney, OCT, M.Sc., B.Ed., B.Sc.

Introduction to MPI. SHARCNET MPI Lecture Series: Part I of II. Paul Preney, OCT, M.Sc., B.Ed., B.Sc. Introduction to MPI SHARCNET MPI Lecture Series: Part I of II Paul Preney, OCT, M.Sc., B.Ed., B.Sc. preney@sharcnet.ca School of Computer Science University of Windsor Windsor, Ontario, Canada Copyright

More information

More about MPI programming. More about MPI programming p. 1

More about MPI programming. More about MPI programming p. 1 More about MPI programming More about MPI programming p. 1 Some recaps (1) One way of categorizing parallel computers is by looking at the memory configuration: In shared-memory systems, the CPUs share

More information

Programming with MPI Collectives

Programming with MPI Collectives Programming with MPI Collectives Jan Thorbecke Type to enter text Delft University of Technology Challenge the future Collectives Classes Communication types exercise: BroadcastBarrier Gather Scatter exercise:

More information

MA471. Lecture 5. Collective MPI Communication

MA471. Lecture 5. Collective MPI Communication MA471 Lecture 5 Collective MPI Communication Today: When all the processes want to send, receive or both Excellent website for MPI command syntax available at: http://www-unix.mcs.anl.gov/mpi/www/ 9/10/2003

More information

First day. Basics of parallel programming. RIKEN CCS HPC Summer School Hiroya Matsuba, RIKEN CCS

First day. Basics of parallel programming. RIKEN CCS HPC Summer School Hiroya Matsuba, RIKEN CCS First day Basics of parallel programming RIKEN CCS HPC Summer School Hiroya Matsuba, RIKEN CCS Today s schedule: Basics of parallel programming 7/22 AM: Lecture Goals Understand the design of typical parallel

More information

MPI and OpenMP (Lecture 25, cs262a) Ion Stoica, UC Berkeley November 19, 2016

MPI and OpenMP (Lecture 25, cs262a) Ion Stoica, UC Berkeley November 19, 2016 MPI and OpenMP (Lecture 25, cs262a) Ion Stoica, UC Berkeley November 19, 2016 Message passing vs. Shared memory Client Client Client Client send(msg) recv(msg) send(msg) recv(msg) MSG MSG MSG IPC Shared

More information

MPI 5. CSCI 4850/5850 High-Performance Computing Spring 2018

MPI 5. CSCI 4850/5850 High-Performance Computing Spring 2018 MPI 5 CSCI 4850/5850 High-Performance Computing Spring 2018 Tae-Hyuk (Ted) Ahn Department of Computer Science Program of Bioinformatics and Computational Biology Saint Louis University Learning Objectives

More information

Parallel Programming, MPI Lecture 2

Parallel Programming, MPI Lecture 2 Parallel Programming, MPI Lecture 2 Ehsan Nedaaee Oskoee 1 1 Department of Physics IASBS IPM Grid and HPC workshop IV, 2011 Outline 1 Introduction and Review The Von Neumann Computer Kinds of Parallel

More information

mith College Computer Science CSC352 Week #7 Spring 2017 Introduction to MPI Dominique Thiébaut

mith College Computer Science CSC352 Week #7 Spring 2017 Introduction to MPI Dominique Thiébaut mith College CSC352 Week #7 Spring 2017 Introduction to MPI Dominique Thiébaut dthiebaut@smith.edu Introduction to MPI D. Thiebaut Inspiration Reference MPI by Blaise Barney, Lawrence Livermore National

More information

Distributed Memory Systems: Part IV

Distributed Memory Systems: Part IV Chapter 5 Distributed Memory Systems: Part IV Max Planck Institute Magdeburg Jens Saak, Scientific Computing II 293/342 The Message Passing Interface is a standard for creation of parallel programs using

More information

Parallel Programming, MPI Lecture 2

Parallel Programming, MPI Lecture 2 Parallel Programming, MPI Lecture 2 Ehsan Nedaaee Oskoee 1 1 Department of Physics IASBS IPM Grid and HPC workshop IV, 2011 Outline 1 Point-to-Point Communication Non Blocking PTP Communication 2 Collective

More information

MPI - The Message Passing Interface

MPI - The Message Passing Interface MPI - The Message Passing Interface The Message Passing Interface (MPI) was first standardized in 1994. De facto standard for distributed memory machines. All Top500 machines (http://www.top500.org) are

More information

Lecture 7: Distributed memory

Lecture 7: Distributed memory Lecture 7: Distributed memory David Bindel 15 Feb 2010 Logistics HW 1 due Wednesday: See wiki for notes on: Bottom-up strategy and debugging Matrix allocation issues Using SSE and alignment comments Timing

More information

mpi-02.c 1/1. 15/10/26 mpi-01.c 1/1. 15/10/26

mpi-02.c 1/1. 15/10/26 mpi-01.c 1/1. 15/10/26 mpi-01.c 1/1 main ( argc, char * argv[]) rank, size; prf ("I am process %d of %d\n", rank, size); mpi-02.c 1/1 #include main ( argc, char * argv[]) rank, size, src, dest, nc; tag = 50; // tag

More information

Basic MPI Communications. Basic MPI Communications (cont d)

Basic MPI Communications. Basic MPI Communications (cont d) Basic MPI Communications MPI provides two non-blocking routines: MPI_Isend(buf,cnt,type,dst,tag,comm,reqHandle) buf: source of data to be sent cnt: number of data elements to be sent type: type of each

More information

Standard MPI - Message Passing Interface

Standard MPI - Message Passing Interface c Ewa Szynkiewicz, 2007 1 Standard MPI - Message Passing Interface The message-passing paradigm is one of the oldest and most widely used approaches for programming parallel machines, especially those

More information

Programming Scalable Systems with MPI. Clemens Grelck, University of Amsterdam

Programming Scalable Systems with MPI. Clemens Grelck, University of Amsterdam Clemens Grelck University of Amsterdam UvA / SurfSARA High Performance Computing and Big Data Course June 2014 Parallel Programming with Compiler Directives: OpenMP Message Passing Gentle Introduction

More information

Parallel Programming. Using MPI (Message Passing Interface)

Parallel Programming. Using MPI (Message Passing Interface) Parallel Programming Using MPI (Message Passing Interface) Message Passing Model Simple implementation of the task/channel model Task Process Channel Message Suitable for a multicomputer Number of processes

More information

Lecture 6: Message Passing Interface

Lecture 6: Message Passing Interface Lecture 6: Message Passing Interface Introduction The basics of MPI Some simple problems More advanced functions of MPI A few more examples CA463D Lecture Notes (Martin Crane 2013) 50 When is Parallel

More information

Practical Introduction to Message-Passing Interface (MPI)

Practical Introduction to Message-Passing Interface (MPI) 1 Practical Introduction to Message-Passing Interface (MPI) October 1st, 2015 By: Pier-Luc St-Onge Partners and Sponsors 2 Setup for the workshop 1. Get a user ID and password paper (provided in class):

More information

CSE 160 Lecture 18. Message Passing

CSE 160 Lecture 18. Message Passing CSE 160 Lecture 18 Message Passing Question 4c % Serial Loop: for i = 1:n/3-1 x(2*i) = x(3*i); % Restructured for Parallelism (CORRECT) for i = 1:3:n/3-1 y(2*i) = y(3*i); for i = 2:3:n/3-1 y(2*i) = y(3*i);

More information

Parallel Computing. Distributed memory model MPI. Leopold Grinberg T. J. Watson IBM Research Center, USA. Instructor: Leopold Grinberg

Parallel Computing. Distributed memory model MPI. Leopold Grinberg T. J. Watson IBM Research Center, USA. Instructor: Leopold Grinberg Parallel Computing Distributed memory model MPI Leopold Grinberg T. J. Watson IBM Research Center, USA Why do we need to compute in parallel large problem size - memory constraints computation on a single

More information

NUMERICAL PARALLEL COMPUTING

NUMERICAL PARALLEL COMPUTING Lecture 5, March 23, 2012: The Message Passing Interface http://people.inf.ethz.ch/iyves/pnc12/ Peter Arbenz, Andreas Adelmann Computer Science Dept, ETH Zürich E-mail: arbenz@inf.ethz.ch Paul Scherrer

More information

Topics. Lecture 7. Review. Other MPI collective functions. Collective Communication (cont d) MPI Programming (III)

Topics. Lecture 7. Review. Other MPI collective functions. Collective Communication (cont d) MPI Programming (III) Topics Lecture 7 MPI Programming (III) Collective communication (cont d) Point-to-point communication Basic point-to-point communication Non-blocking point-to-point communication Four modes of blocking

More information

int sum;... sum = sum + c?

int sum;... sum = sum + c? int sum;... sum = sum + c? Version Cores Time (secs) Speedup manycore Message Passing Interface mpiexec int main( ) { int ; char ; } MPI_Init( ); MPI_Comm_size(, &N); MPI_Comm_rank(, &R); gethostname(

More information

Data parallelism. [ any app performing the *same* operation across a data stream ]

Data parallelism. [ any app performing the *same* operation across a data stream ] Data parallelism [ any app performing the *same* operation across a data stream ] Contrast stretching: Version Cores Time (secs) Speedup while (step < NumSteps &&!converged) { step++; diffs = 0; foreach

More information

CS4961 Parallel Programming. Lecture 16: Introduction to Message Passing 11/3/11. Administrative. Mary Hall November 3, 2011.

CS4961 Parallel Programming. Lecture 16: Introduction to Message Passing 11/3/11. Administrative. Mary Hall November 3, 2011. CS4961 Parallel Programming Lecture 16: Introduction to Message Passing Administrative Next programming assignment due on Monday, Nov. 7 at midnight Need to define teams and have initial conversation with

More information

High Performance Computing Course Notes Message Passing Programming I

High Performance Computing Course Notes Message Passing Programming I High Performance Computing Course Notes 2008-2009 2009 Message Passing Programming I Message Passing Programming Message Passing is the most widely used parallel programming model Message passing works

More information

Introduction to MPI. SuperComputing Applications and Innovation Department 1 / 143

Introduction to MPI. SuperComputing Applications and Innovation Department 1 / 143 Introduction to MPI Isabella Baccarelli - i.baccarelli@cineca.it Mariella Ippolito - m.ippolito@cineca.it Cristiano Padrin - c.padrin@cineca.it Vittorio Ruggiero - v.ruggiero@cineca.it SuperComputing Applications

More information

CS 470 Spring Mike Lam, Professor. Distributed Programming & MPI

CS 470 Spring Mike Lam, Professor. Distributed Programming & MPI CS 470 Spring 2019 Mike Lam, Professor Distributed Programming & MPI MPI paradigm Single program, multiple data (SPMD) One program, multiple processes (ranks) Processes communicate via messages An MPI

More information

Outline. Introduction to HPC computing. OpenMP MPI. Introduction. Understanding communications. Collective communications. Communicators.

Outline. Introduction to HPC computing. OpenMP MPI. Introduction. Understanding communications. Collective communications. Communicators. Lecture 8 MPI Outline Introduction to HPC computing OpenMP MPI Introduction Understanding communications Collective communications Communicators Topologies Grouping Data for Communication Input / output

More information

Collective Communications II

Collective Communications II Collective Communications II Ned Nedialkov McMaster University Canada SE/CS 4F03 January 2014 Outline Scatter Example: parallel A b Distributing a matrix Gather Serial A b Parallel A b Allocating memory

More information

MPI point-to-point communication

MPI point-to-point communication MPI point-to-point communication Slides Sebastian von Alfthan CSC Tieteen tietotekniikan keskus Oy CSC IT Center for Science Ltd. Introduction MPI processes are independent, they communicate to coordinate

More information

MPI MESSAGE PASSING INTERFACE

MPI MESSAGE PASSING INTERFACE MPI MESSAGE PASSING INTERFACE David COLIGNON, ULiège CÉCI - Consortium des Équipements de Calcul Intensif http://www.ceci-hpc.be Outline Introduction From serial source code to parallel execution MPI functions

More information

Distributed Memory Programming with MPI

Distributed Memory Programming with MPI Distributed Memory Programming with MPI Part 1 Bryan Mills, PhD Spring 2017 A distributed memory system A shared memory system Identifying MPI processes n Common pracace to idenafy processes by nonnegaave

More information

COSC 6374 Parallel Computation. Message Passing Interface (MPI ) I Introduction. Distributed memory machines

COSC 6374 Parallel Computation. Message Passing Interface (MPI ) I Introduction. Distributed memory machines Network card Network card 1 COSC 6374 Parallel Computation Message Passing Interface (MPI ) I Introduction Edgar Gabriel Fall 015 Distributed memory machines Each compute node represents an independent

More information

The Message Passing Interface (MPI) TMA4280 Introduction to Supercomputing

The Message Passing Interface (MPI) TMA4280 Introduction to Supercomputing The Message Passing Interface (MPI) TMA4280 Introduction to Supercomputing NTNU, IMF January 16. 2017 1 Parallelism Decompose the execution into several tasks according to the work to be done: Function/Task

More information

MPI Tutorial. Shao-Ching Huang. High Performance Computing Group UCLA Institute for Digital Research and Education

MPI Tutorial. Shao-Ching Huang. High Performance Computing Group UCLA Institute for Digital Research and Education MPI Tutorial Shao-Ching Huang High Performance Computing Group UCLA Institute for Digital Research and Education Center for Vision, Cognition, Learning and Art, UCLA July 15 22, 2013 A few words before

More information

Message Passing Interface - MPI

Message Passing Interface - MPI Message Passing Interface - MPI Parallel and Distributed Computing Department of Computer Science and Engineering (DEI) Instituto Superior Técnico October 24, 2011 Many slides adapted from lectures by

More information

Advanced MPI. Andrew Emerson

Advanced MPI. Andrew Emerson Advanced MPI Andrew Emerson (a.emerson@cineca.it) Agenda 1. One sided Communications (MPI-2) 2. Dynamic processes (MPI-2) 3. Profiling MPI and tracing 4. MPI-I/O 5. MPI-3 11/12/2015 Advanced MPI 2 One

More information

Parallel Computing Paradigms

Parallel Computing Paradigms Parallel Computing Paradigms Message Passing João Luís Ferreira Sobral Departamento do Informática Universidade do Minho 31 October 2017 Communication paradigms for distributed memory Message passing is

More information

What is Hadoop? Hadoop is an ecosystem of tools for processing Big Data. Hadoop is an open source project.

What is Hadoop? Hadoop is an ecosystem of tools for processing Big Data. Hadoop is an open source project. Back to Hadoop 1 What is Hadoop? Hadoop is an ecosystem of tools for processing Big Data. Hadoop is an open source project. 2 A family of tools MapReduce HDFS HBase Hive Pig ZooKeeper Avro Sqoop Oozie

More information

Scientific Computing

Scientific Computing Lecture on Scientific Computing Dr. Kersten Schmidt Lecture 21 Technische Universität Berlin Institut für Mathematik Wintersemester 2014/2015 Syllabus Linear Regression, Fast Fourier transform Modelling

More information

Programming with MPI. Pedro Velho

Programming with MPI. Pedro Velho Programming with MPI Pedro Velho Science Research Challenges Some applications require tremendous computing power - Stress the limits of computing power and storage - Who might be interested in those applications?

More information

Introduction to MPI: Part II

Introduction to MPI: Part II Introduction to MPI: Part II Pawel Pomorski, University of Waterloo, SHARCNET ppomorsk@sharcnetca November 25, 2015 Summary of Part I: To write working MPI (Message Passing Interface) parallel programs

More information

Collective Communications I

Collective Communications I Collective Communications I Ned Nedialkov McMaster University Canada CS/SE 4F03 January 2016 Outline Introduction Broadcast Reduce c 2013 16 Ned Nedialkov 2/14 Introduction A collective communication involves

More information

MPI Tutorial. Shao-Ching Huang. IDRE High Performance Computing Workshop

MPI Tutorial. Shao-Ching Huang. IDRE High Performance Computing Workshop MPI Tutorial Shao-Ching Huang IDRE High Performance Computing Workshop 2013-02-13 Distributed Memory Each CPU has its own (local) memory This needs to be fast for parallel scalability (e.g. Infiniband,

More information

CEE 618 Scientific Parallel Computing (Lecture 5): Message-Passing Interface (MPI) advanced

CEE 618 Scientific Parallel Computing (Lecture 5): Message-Passing Interface (MPI) advanced 1 / 32 CEE 618 Scientific Parallel Computing (Lecture 5): Message-Passing Interface (MPI) advanced Albert S. Kim Department of Civil and Environmental Engineering University of Hawai i at Manoa 2540 Dole

More information

Department of Informatics V. HPC-Lab. Session 4: MPI, CG M. Bader, A. Breuer. Alex Breuer

Department of Informatics V. HPC-Lab. Session 4: MPI, CG M. Bader, A. Breuer. Alex Breuer HPC-Lab Session 4: MPI, CG M. Bader, A. Breuer Meetings Date Schedule 10/13/14 Kickoff 10/20/14 Q&A 10/27/14 Presentation 1 11/03/14 H. Bast, Intel 11/10/14 Presentation 2 12/01/14 Presentation 3 12/08/14

More information

Parallel hardware. Distributed Memory. Parallel software. COMP528 MPI Programming, I. Flynn s taxonomy:

Parallel hardware. Distributed Memory. Parallel software. COMP528 MPI Programming, I. Flynn s taxonomy: COMP528 MPI Programming, I www.csc.liv.ac.uk/~alexei/comp528 Alexei Lisitsa Dept of computer science University of Liverpool a.lisitsa@.liverpool.ac.uk Flynn s taxonomy: Parallel hardware SISD (Single

More information

The Message Passing Interface (MPI): Parallelism on Multiple (Possibly Heterogeneous) CPUs

The Message Passing Interface (MPI): Parallelism on Multiple (Possibly Heterogeneous) CPUs 1 The Message Passing Interface (MPI): Parallelism on Multiple (Possibly Heterogeneous) CPUs http://mpi-forum.org https://www.open-mpi.org/ Mike Bailey mjb@cs.oregonstate.edu Oregon State University mpi.pptx

More information

Capstone Project. Project: Middleware for Cluster Computing

Capstone Project. Project: Middleware for Cluster Computing Capstone Project Project: Middleware for Cluster Computing Middleware is computer software that connects software components or applications. The software consists of a set of enabling services that allow

More information

Distributed Memory Parallel Programming

Distributed Memory Parallel Programming COSC Big Data Analytics Parallel Programming using MPI Edgar Gabriel Spring 201 Distributed Memory Parallel Programming Vast majority of clusters are homogeneous Necessitated by the complexity of maintaining

More information

15-440: Recitation 8

15-440: Recitation 8 15-440: Recitation 8 School of Computer Science Carnegie Mellon University, Qatar Fall 2013 Date: Oct 31, 2013 I- Intended Learning Outcome (ILO): The ILO of this recitation is: Apply parallel programs

More information

High-Performance Computing: MPI (ctd)

High-Performance Computing: MPI (ctd) High-Performance Computing: MPI (ctd) Adrian F. Clark: alien@essex.ac.uk 2015 16 Adrian F. Clark: alien@essex.ac.uk High-Performance Computing: MPI (ctd) 2015 16 1 / 22 A reminder Last time, we started

More information

Programming Scalable Systems with MPI. UvA / SURFsara High Performance Computing and Big Data. Clemens Grelck, University of Amsterdam

Programming Scalable Systems with MPI. UvA / SURFsara High Performance Computing and Big Data. Clemens Grelck, University of Amsterdam Clemens Grelck University of Amsterdam UvA / SURFsara High Performance Computing and Big Data Message Passing as a Programming Paradigm Gentle Introduction to MPI Point-to-point Communication Message Passing

More information

MPI: The Message-Passing Interface. Most of this discussion is from [1] and [2].

MPI: The Message-Passing Interface. Most of this discussion is from [1] and [2]. MPI: The Message-Passing Interface Most of this discussion is from [1] and [2]. What Is MPI? The Message-Passing Interface (MPI) is a standard for expressing distributed parallelism via message passing.

More information

Introduction to Parallel Programming

Introduction to Parallel Programming University of Nizhni Novgorod Faculty of Computational Mathematics & Cybernetics Section 4. Part 1. Introduction to Parallel Programming Parallel Programming with MPI Gergel V.P., Professor, D.Sc., Software

More information

MPI introduction - exercises -

MPI introduction - exercises - MPI introduction - exercises - Paolo Ramieri, Maurizio Cremonesi May 2016 Startup notes Access the server and go on scratch partition: ssh a08tra49@login.galileo.cineca.it cd $CINECA_SCRATCH Create a job

More information

What s in this talk? Quick Introduction. Programming in Parallel

What s in this talk? Quick Introduction. Programming in Parallel What s in this talk? Parallel programming methodologies - why MPI? Where can I use MPI? MPI in action Getting MPI to work at Warwick Examples MPI: Parallel Programming for Extreme Machines Si Hammond,

More information