CRESCO Project: Salvatore Raia

Slides:



Advertisements
Similar presentations
© 2010 UC Regents | UC Berkeley ist.berkeley.edu IST Platform Services IST Platform Services and Web Hosting Options Micronet Meeting June 23, 2010 Don.
Advertisements

INFO Direzione INFO Coordinatore Calcolo Scientifico Ing. Silvio Migliori (5-2003)
PARMON A Comprehensive Cluster Monitoring System PARMON Team Centre for Development of Advanced Computing, Bangalore, India Contact: Rajkumar Buyya
Issues of HPC software From the experience of TH-1A Lu Yutong NUDT.
Setting up of condor scheduler on computing cluster Raman Sehgal NPD-BARC.
ENEA-GRID and gLite Interoperability: robustness of SPAGO approach Catania, Italy, February ENEA-GRID and gLite Interoperability: robustness.
IBM RS/6000 SP POWER3 SMP Jari Jokinen Pekka Laurila.
HPCC Mid-Morning Break Interactive High Performance Computing Dirk Colbry, Ph.D. Research Specialist Institute for Cyber Enabled Discovery.
Measuring zSeries System Performance Dr. Chu J. Jong School of Information Technology Illinois State University 06/11/2012 Sponsored in part by Deer &
Introduction to Symmetric Multiprocessors Süha TUNA Bilişim Enstitüsü UHeM Yaz Çalıştayı
Cloud Computing for the Enterprise November 18th, This work is licensed under a Creative Commons.
Reference: / Parallel Programming Paradigm Yeni Herdiyeni Dept of Computer Science, IPB.
Parallel Processing LAB NO 1.
Hands-On Microsoft Windows Server 2008 Chapter 1 Introduction to Windows Server 2008.
 What is an operating system? What is an operating system?  Where does the OS fit in? Where does the OS fit in?  Services provided by an OS Services.
Cloud Computing 1. Outline  Introduction  Evolution  Cloud architecture  Map reduce operation  Platform 2.
Tools and Utilities for parallel and serial codes in ENEA-GRID environment CRESCO Project: Salvatore Raia SubProject I.2 C.R. ENEA-Portici. 11/12/2007.
Running Climate Models On The NERC Cluster Grid Using G-Rex Dan Bretherton, Jon Blower and Keith Haines Reading e-Science Centre Environmental.
University of Illinois at Urbana-Champaign NCSA Supercluster Administration NT Cluster Group Computing and Communications Division NCSA Avneesh Pant
SA1 / Operation & support Enabling Grids for E-sciencE Integration of heterogeneous computational resources in.
Seaborg Cerise Wuthrich CMPS Seaborg  Manufactured by IBM  Distributed Memory Parallel Supercomputer  Based on IBM’s SP RS/6000 Architecture.
March 3rd, 2006 Chen Peng, Lilly System Biology1 Cluster and SGE.
Trace Generation to Simulate Large Scale Distributed Application Olivier Dalle, Emiio P. ManciniMar. 8th, 2012.
Enabling Grids for E-sciencE CRESCO HPC SYSTEM INTEGRATED INTO ENEA GRID ENVIRONMENT G. Bracco, S.
SUMA: A Scientific Metacomputer Cardinale, Yudith Figueira, Carlos Hernández, Emilio Baquero, Eduardo Berbín, Luis Bouza, Roberto Gamess, Eric García,
INTRODUCTION SOFTWARE HARDWARE DIFFERENCE BETWEEN THE S/W AND H/W.
Tests and tools for ENEA GRID Performance test: HPL (High Performance Linpack) Network monitoring A.Funel December 11, 2007.
Tool Integration with Data and Computation Grid GWE - “Grid Wizard Enterprise”
ARGONNE NATIONAL LABORATORY Climate Modeling on the Jazz Linux Cluster at ANL John Taylor Mathematics and Computer Science & Environmental Research Divisions.
Issues Autonomic operation (fault tolerance) Minimize interference to applications Hardware support for new operating systems Resource management (global.
Nanco: a large HPC cluster for RBNI (Russell Berrie Nanotechnology Institute) Anne Weill – Zrahia Technion,Computer Center October 2008.
Enabling Grids for E-sciencE CRESCO COMPUTATIONAL RESOURCES AND ITS INTEGRATION IN ENEA-GRID.
How to for compiling and running MPI Programs. Prepared by Kiriti Venkat.
Data Management for Decision Support Session-4 Prof. Bharat Bhasker.
COMP381 by M. Hamdi 1 Clusters: Networks of WS/PC.
Tool Integration with Data and Computation Grid “Grid Wizard 2”
Chapter 1 Basic Concepts of Operating Systems Introduction Software A program is a sequence of instructions that enables the computer to carry.
Integration of heterogeneous computational resources in EGEE: a live demo Istanbul, Turkey, September Integration of heterogeneous computational.
Grid Remote Execution of Large Climate Models (NERC Cluster Grid) Dan Bretherton, Jon Blower and Keith Haines Reading e-Science Centre
CIP HPC CIP - HPC HPC = High Performance Computer It’s not a regular computer, it’s bigger, faster, more powerful, and more.
Background Computer System Architectures Computer System Software.
CLIENT SERVER COMPUTING. We have 2 types of n/w architectures – client server and peer to peer. In P2P, each system has equal capabilities and responsibilities.
Hybrid Parallel Implementation of The DG Method Advanced Computing Department/ CAAM 03/03/2016 N. Chaabane, B. Riviere, H. Calandra, M. Sekachev, S. Hamlaoui.
POE Parallel Operating Environment. Cliff Montgomery.
CNAF - 24 September 2004 EGEE SA-1 SPACI Activity Italo Epicoco.
INTRODUCTION TO HIGH PERFORMANCE COMPUTING AND TERMINOLOGY.
Introduction to Parallel Computing: MPI, OpenMP and Hybrid Programming
GWE Core Grid Wizard Enterprise (
Consulting Services JobScheduler Architecture Decision Template
Porting MM5 and BOLAM codes to the GRID
CompChem VO: User experience using MPI
NGS computation services: APIs and Parallel Jobs
Is System X for Me? Cal Ribbens Computer Science Department
TYPES OFF OPERATING SYSTEM
NCSA Supercluster Administration
Chapter 17: Database System Architectures
Introduction to Apache
Support for ”interactive batch”
CLUSTER COMPUTING.
Constructing a system with multiple computers or processors
Hybrid Programming with OpenMP and MPI
EFDA Meeting – Portici ITALIAN NATIONAL AGENCY
MPJ: A Java-based Parallel Computing System
High Performance Computing
So what is Target Management all about?
Introduction to High Performance Computing Using Sapelo2 at GACRC
Chapter 4 Multiprocessors
A Virtual Machine Monitor for Utilizing Non-dedicated Clusters
Working in The IITJ HPC System
Presentation transcript:

CRESCO Project: Salvatore Raia C.R. ENEA-Portici. 11/12/2007 Tools and Utilities for parallel and serial codes in ENEA-GRID environment CRESCO Project: Salvatore Raia SubProject I.2

OUTLINE: GRID, cluster and parallel Computing (Intro) ENEA-GRID. Architecture and functionality My Activity for CRESCO project and results on ENEA-GRID Conclusion and objectives C.R. ENEA-Portici. 11/12/2007

What is a cluster ? Collection of resources (HW, SW) connected via public or private network - Each CPU runs a separated istance of operating system -Administration: local Supercomputer= computer with many processors connected via high-speed computer bus and that share the memory (SMP) . It runs one Operating system cluster 1 supercomputer C.R. ENEA-Portici. 11/12/2007

GRID = nodes made of clusters and each node may have Shared or How to get a Grid ? Collection of interconnected clusters geographically distributed - administration: sometimes clusters belong to different department or company GRID = nodes made of clusters and each node may have Shared or Distributed memory architectures (Hybrid ) that share processes . ENEA-GRID has the same structure With 6 clusters: Bologna, Casaccia, Frascati, Portici, Trisaia, Brindisi cluster N cluster 3 cluster 2 cluster 1 GRID 1 C.R. ENEA-Portici. 11/12/2007

ENEA-GRID structure (HW) C.R. ENEA-Portici. 11/12/2007

GRID features Pro: Con: Shared resources Low costs (clock ?) Frequency scaling (domain ?) Power consumption P=C×V×V×F Pro: Shared resources Low costs (clock ?) Open systems Scalability Con: Several platforms Load balancing User Access How is it managed on ENEA-GRID ? C.R. ENEA-Portici. 11/12/2007

ENEA-GRID structure (SW) Resources management ICA client File System Operating Systems C.R. ENEA-Portici. 11/12/2007

User Interface Switch host USER ACCESS ICA client ssh o telnet web Run Appl. Jobs status USER ACCESS ICA client ssh o telnet web C.R. ENEA-Portici. 11/12/2007

My activity on ENEA-GRID (CRESCO pr.) Serial and Parallel (MPI) codes How to cope with ? Problem with: Multi platforms Load balancing User Access User interfaces LSF utilities Software dev. C.R. ENEA-Portici. 11/12/2007

Tools for Serial and Parallel (MPI) codes Multi Platform …So we need a lots of binaries for each platform. Launcher: after compiling our source code in each platform, we have “binary1”…”binaryN” for host1,…hostN. It is a shell script (placed on AFS) that selects the righteous “binary” for the selected host Serial codes Compilers GNU PGI IBM Parallel codes (MPI) MPI Implementations MPICH LAM-MPI POE Problems with execution too …tools C.R. ENEA-Portici. 11/12/2007

Some MPI problems C.R. ENEA-Portici. 11/12/2007

Results: tools serial and parallel (MPI) codes Program for Fortran 77/90,C and C++ serial compiling (look Java Interface) Launcher for “NS2” application (use external libraries) PARALLEL (MPI) Launcher for running a test program (check command) Launcher for HPL test on AIX and Linux user1 installation user2 installation C.R. ENEA-Portici. 11/12/2007

Analizing LSF utilities Serial and Parallel codes LSF Resources Serial codes Resources definition “NS2” application Serial LSF utilities Job array (Multicase) “lsgrun” Parallel codes (MPI) Parallel LSF utilities “mpijob” (MPICH) “poejob” (POE) Correlation No correlation C.R. ENEA-Portici. 11/12/2007

Results: Integration with other application (My)Java Interface Serial codes Parallel codes (MPI) C.R. ENEA-Portici. 11/12/2007

Conclusion and objectives Launcher + LSF utilities + User interface allow to create a omogeneous environment Objectives: Optimization of programs to launch serial and parallel codes, including checking resources to run the application (e.g. library, other programs, etc) Exploitation of LSF utilities in order to make easy running MPI programs (mpijob, poejob, etc) and load balancing Improve error handling for user interfaces … … C.R. ENEA-Portici. 11/12/2007

Andrew File System C.R. ENEA-Portici. 11/12/2007

LSF-Load Sharing Facilities C.R. ENEA-Portici. 11/12/2007