Rocks Clusters SUN HPC Consortium November 2004 Federico D. Sacerdoti Advanced CyberInfrastructure Group San Diego Supercomputer Center.

Slides:



Advertisements
Similar presentations
Kejun Dong, Kai Nan CNIC/CAS CNIC Resources And Activities Update Resources Working Group PRAGMA11 Workshop, Oct.16/17 Osaka, Japan.
Advertisements

Test harness and reporting framework Shava Smallen San Diego Supercomputer Center Grid Performance Workshop 6/22/05.
1 Applications Virtualization in VPC Nadya Williams UCSD.
1 Agenda … HPC Technology & Trends HPC Platforms & Roadmaps HP Supercomputing Vision HP Today.
Windows Deployment Services WDS for Large Scale Enterprises and Small IT Shops Presented By: Ryan Drown Systems Administrator for Krannert.
High Performance Linux Clusters Guru Session, Usenix, Boston June 30, 2004 Greg Bruno, SDSC.
2. Computer Clusters for Scalable Parallel Computing
Beowulf Supercomputer System Lee, Jung won CS843.
Quick Overview of NPACI Rocks Philip M. Papadopoulos Associate Director, Distributed Computing San Diego Supercomputer Center.
Managing Linux Clusters with Rocks Tim Carlson - PNNL
Leveraging WinPE and Linux Preboot for Effective Provisioning Jonathan Richey | Director of Development | Altiris, Inc.
© UC Regents 2010 Extending Rocks Clusters into Amazon EC2 Using Condor Philip Papadopoulos, Ph.D University of California, San Diego San Diego Supercomputer.
Building a Cluster Support Service Implementation of the SCS Program UC Computing Services Conference Gary Jung SCS Project Manager
Linux clustering Morris Law, IT Coordinator, Science Faculty, Hong Kong Baptist University.
Chiba City: A Testbed for Scalablity and Development FAST-OS Workshop July 10, 2002 Rémy Evard Mathematics.
CPP Staff - 30 CPP Staff - 30 FCIPT Staff - 35 IPR Staff IPR Staff ITER-India Staff ITER-India Staff Research Areas: 1.Studies.
Cluster computing facility for CMS simulation work at NPD-BARC Raman Sehgal.
Building a High-performance Computing Cluster Using FreeBSD BSDCon '03 September 10, 2003 Brooks Davis, Michael AuYeung, Gary Green, Craig Lee The Aerospace.
THE AFFORDABLE SUPERCOMPUTER HARRISON CARRANZA APARICIO CARRANZA JOSE REYES ALAMO CUNY – NEW YORK CITY COLLEGE OF TECHNOLOGY ECC Conference 2015 – June.
Rocks cluster : a cluster oriented linux distribution or how to install a computer cluster in a day.
05/18/03Maurizio Davini Hepix2003 Department of Physics University of Pisa Site Report Maurizio Davini Department of Physics and INFN Pisa.
SSI-OSCAR A Single System Image for OSCAR Clusters Geoffroy Vallée INRIA – PARIS project team COSET-1 June 26th, 2004.
Track 1: Cluster and Grid Computing NBCR Summer Institute Session 2.2: Cluster and Grid Computing: Case studies Condor introduction August 9, 2006 Nadya.
Virtualization Lab 3 – Virtualization Fall 2012 CSCI 6303 Principles of I.T.
ROCKS & The CASCI Cluster By Rick Bohn. What’s a Cluster? Cluster is a widely-used term meaning independent computers combined into a unified system through.
National Center for Supercomputing Applications The Computational Chemistry Grid: Production Cyberinfrastructure for Computational Chemistry PI: John Connolly.
© 2006 UC Regents1 Track 1: Cluster and Grid Computing NBCR Summer Institute Session 1.1: Introduction to Cluster and Grid Computing August 8, 2006 Nadya.
So, Jung-ki Distributed Computing System LAB School of Computer Science and Engineering Seoul National University Implementation of Package Management.
October, Scientific Linux INFN/Trieste B.Gobbo – Compass R.Gomezel - T.Macorini - L.Strizzolo INFN - Trieste.
Rocks ‘n’ Rolls An Introduction to Programming Clusters using Rocks © 2008 UC Regents Anoop Rajendra.
การติดตั้งและทดสอบการทำคลัสเต อร์เสมือนบน Xen, ROCKS, และไท ยกริด Roll Implementation of Virtualization Clusters based on Xen, ROCKS, and ThaiGrid Roll.
Taking the Complexity out of Cluster Computing Vendor Update HPC User Forum Arend Dittmer Director Product Management HPC April,
Batch Scheduling at LeSC with Sun Grid Engine David McBride Systems Programmer London e-Science Centre Department of Computing, Imperial College.
Grid MP at ISIS Tom Griffin, ISIS Facility. Introduction About ISIS Why Grid MP? About Grid MP Examples The future.
Computer Systems Lab The University of Wisconsin - Madison Department of Computer Sciences Linux Clusters David Thompson
SoCal Infrastructure OptIPuter Southern California Network Infrastructure Philip Papadopoulos OptIPuter Co-PI University of California, San Diego Program.
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.
© 2007 UC Regents1 Track 1: Cluster and Grid Computing NBCR Summer Institute Session 1.1: Introduction to Cluster and Grid Computing July 31, 2007 Wilfred.
Deploying a Network of GNU/Linux Clusters with Rocks / Arto Teräs Slide 1(18) Deploying a Network of GNU/Linux Clusters with Rocks Arto Teräs.
NA-MIC National Alliance for Medical Image Computing UCSD: Engineering Core 2 Portal and Grid Infrastructure.
1October 9, 2001 Sun in Scientific & Engineering Computing Grid Computing with Sun Wolfgang Gentzsch Director Grid Computing Cracow Grid Workshop, November.
Installing, running, and maintaining large Linux Clusters at CERN Thorsten Kleinwort CERN-IT/FIO CHEP
GRIDS Center Middleware Overview Sandra Redman Information Technology and Systems Center and Information Technology Research Center National Space Science.
Vicky Rowley Solution Architect BIRN Coordinating Center - University of California San Diego E-x-t-e-n-d-i-n-g Rocks: The Creation and Management of Grid.
VMWare Workstation Installation. Starting Vmware Workstation Go to the start menu and start the VMware Workstation program. *Note: The following instructions.
EVGM081 Multi-Site Virtual Cluster: A User-Oriented, Distributed Deployment and Management Mechanism for Grid Computing Environments Takahiro Hirofuchi,
GCRC Meeting 2004 BIRN Coordinating Center Software Development Vicky Rowley.
Cluster Software Overview
PerfSONAR-PS Functionality February 11 th 2010, APAN 29 – perfSONAR Workshop Jeff Boote, Assistant Director R&D.
HEP Computing Status Sheffield University Matt Robinson Paul Hodgson Andrew Beresford.
1 Copyright © 2015 Pexus LLC Patriot PS Personal Server Installing Patriot PS ISO Image on.
Internet2 AdvCollab Apps 1 Access Grid Vision To create virtual spaces where distributed people can work together. Challenges:
LCG LCG-1 Deployment and usage experience Lev Shamardin SINP MSU, Moscow
COMP381 by M. Hamdi 1 Clusters: Networks of WS/PC.
Office of Science U.S. Department of Energy NERSC Site Report HEPiX October 20, 2003 TRIUMF.
Tool Integration with Data and Computation Grid “Grid Wizard 2”
Building and managing production bioclusters Chris Dagdigian BIOSILICO Vol2, No. 5 September 2004 Ankur Dhanik.
15-Feb-02Steve Traylen, RAL WP6 Test Bed Report1 RAL/UK WP6 Test Bed Report Steve Traylen, WP6 PPGRID/RAL, UK
R. Krempaska, October, 2013 Wir schaffen Wissen – heute für morgen Controls Security at PSI Current Status R. Krempaska, A. Bertrand, C. Higgs, R. Kapeller,
Automating Installations by Using the Microsoft Windows 2000 Setup Manager Create setup scripts simply and easily. Create and modify answer files and UDFs.
Improving the Research Bootstrap of Condor High Throughput Computing for Non-Cluster Experts Based on Knoppix Instant Computing Technology RIKEN Genomic.
Overview of cluster management tools Marco Mambelli – August OSG Summer Workshop TTU - Lubbock, TX THE UNIVERSITY OF CHICAGO.
OSG Facility Miron Livny OSG Facility Coordinator and PI University of Wisconsin-Madison Open Science Grid Scientific Advisory Group Meeting June 12th.
© 2007 UC Regents1 Rocks – Present and Future The State of Things Open Source Grids and Clusters Conference Philip Papadopoulos, Greg Bruno Mason Katz,
Advanced Network Administration Computer Clusters.
Create setup scripts simply and easily.
GWE Core Grid Wizard Enterprise (
Instructor: Mort Anvari
SCCM in hybrid world Predrag Jelesijević Microsoft 7/6/ :17 AM
Presentation transcript:

Rocks Clusters SUN HPC Consortium November 2004 Federico D. Sacerdoti Advanced CyberInfrastructure Group San Diego Supercomputer Center

Copyright © 2004 F. Sacerdoti, M. Katz, G. Bruno, P. Papadopoulos, UC Regents Outline Rocks Identity Rocks Mission Why Rocks Rocks Design Rocks Technologies, Services, Capabilities Rockstar

Copyright © 2004 F. Sacerdoti, M. Katz, G. Bruno, P. Papadopoulos, UC Regents Rocks Identity System to build and manage Linux Clusters  General Linux maintenance system for N nodes  Desktops too  Happens to be good for clusters Free Mature High Performance  Designed for scientific workloads

Copyright © 2004 F. Sacerdoti, M. Katz, G. Bruno, P. Papadopoulos, UC Regents Rocks Mission Make Clusters Easy (Papadopoulos, 00) Most cluster projects assume a sysadmin will help build the cluster. Build a cluster without assuming CS knowledge  Simple idea, complex ramifications  Automatic configuration of all components and services  ~30 services on frontend, ~10 services on compute nodes  Clusters for Scientists Results in a very robust system that is insulated from human mistakes

Copyright © 2004 F. Sacerdoti, M. Katz, G. Bruno, P. Papadopoulos, UC Regents Why Rocks Easiest way to build a Rockstar-class machine with SGE ready out of the box More supported architectures  Pentium, Athlon, Opteron, Nocona, Itanium More happy users  280 registered clusters, 700 member support list  HPCwire Readers Choice Awards 2004 More configured HPC software: 15 optional extensions (rolls) and counting. Unmatched Release Quality.

Copyright © 2004 F. Sacerdoti, M. Katz, G. Bruno, P. Papadopoulos, UC Regents Why Rocks Big projects use Rocks  BIRN (20 clusters)  GEON (20 clusters)  NBCR (6 clusters) Supports different clustering toolkits  Rocks Standard (RedHat HPC)  SCE  SCore (Single Process Space)  OpenMosix (Single Process Space: on the way)

Copyright © 2004 F. Sacerdoti, M. Katz, G. Bruno, P. Papadopoulos, UC Regents Rocks Design Uses RedHat’s intelligent installer  Leverages RedHat’s ability to discover & configure hardware  Everyone tries System Imaging at first  Who has homogeneous hardware?  If so, whose cluster stays that way? Description Based install: Kickstart  Like Jumpstart Contains a viable Operating System  No need to “pre-configure” an OS

Copyright © 2004 F. Sacerdoti, M. Katz, G. Bruno, P. Papadopoulos, UC Regents Rocks Design No special “Rocksified” package structure. Can install any RPM. Where Linux core packages come from:  RedHat Advanced Workstation (from SRPMS)  Enterprise Linux 3

Copyright © 2004 F. Sacerdoti, M. Katz, G. Bruno, P. Papadopoulos, UC Regents Rocks Leap of Faith Install is primitive operation for Upgrade and Patch  Seems wrong at first  Why must you reinstall the whole thing?  Actually right: debugging a Linux system is fruitless at this scale. Reinstall enforces stability.  Primary user has no sysadmin to help troubleshoot Rocks install is scalable and fast: 15min for entire cluster  Post script work done in parallel by compute nodes. Power Admins may use up2date or yum for patches.  To compute nodes by reinstall

Rocks Technology

Copyright © 2004 F. Sacerdoti, M. Katz, G. Bruno, P. Papadopoulos, UC Regents Cluster Integration with Rocks 1. Build a frontend node 1. Insert CDs: Base, HPC, Kernel, optional Rolls 2. Answer install screens: network, timezone, password 2. Build compute nodes 1. Run insert-ethers on frontend (dhcpd listener) 2. PXE boot compute nodes in name order 3. Start Computing

Copyright © 2004 F. Sacerdoti, M. Katz, G. Bruno, P. Papadopoulos, UC Regents Rocks Tech: Dynamic Kickstart File On node install

Copyright © 2004 F. Sacerdoti, M. Katz, G. Bruno, P. Papadopoulos, UC Regents Rocks Roll Architecture Rolls are Rocks Modules  Think Apache Software for cluster  Packaged  3rd party tarballs  Tested  Automatically configured services RPMS plus Kickstart graph in ISO form.

Copyright © 2004 F. Sacerdoti, M. Katz, G. Bruno, P. Papadopoulos, UC Regents Rocks Tech: Dynamic Kickstart File With Roll (HPC) HPCbase

Copyright © 2004 F. Sacerdoti, M. Katz, G. Bruno, P. Papadopoulos, UC Regents Rocks Tech: Wide Area Net Install Install a frontend without CDs Benefits Can install from minimal boot image Rolls downloaded dynamically Community can build specific extensions

Copyright © 2004 F. Sacerdoti, M. Katz, G. Bruno, P. Papadopoulos, UC Regents Rocks Tech: Security & Encryption To protect the kickstart file

Copyright © 2004 F. Sacerdoti, M. Katz, G. Bruno, P. Papadopoulos, UC Regents Rocks Tech: 411 Information Service 411 does NIS  Distribute passwords File based, simple  HTTP transport  Multicast Scalable Secure

Rocks Services

Copyright © 2004 F. Sacerdoti, M. Katz, G. Bruno, P. Papadopoulos, UC Regents Rocks Cluster Homepage

Copyright © 2004 F. Sacerdoti, M. Katz, G. Bruno, P. Papadopoulos, UC Regents Rocks Services: Ganglia Monitoring

Copyright © 2004 F. Sacerdoti, M. Katz, G. Bruno, P. Papadopoulos, UC Regents Rocks Services: Job Monitoring SGE Batch System

Copyright © 2004 F. Sacerdoti, M. Katz, G. Bruno, P. Papadopoulos, UC Regents Rocks Services: Job Monitoring How a job affects resources on this node

Copyright © 2004 F. Sacerdoti, M. Katz, G. Bruno, P. Papadopoulos, UC Regents Rocks Services: Configured, Ready Grid (Globus, from NMI) Condor (NMI)  Globus GRAM SGE  Globus GRAM MPD parallel job launcher (Argonne)  MPICH 1, 2 Intel Compiler set PVFS

Rocks Capabilities

Copyright © 2004 F. Sacerdoti, M. Katz, G. Bruno, P. Papadopoulos, UC Regents High Performance Interconnect Support Myrinet  All major versions, GM2  Automatic configuration and support in Rocks since first release Infiniband  Via Collaboration with AMD & Infinicon  IB  IPoIB

Copyright © 2004 F. Sacerdoti, M. Katz, G. Bruno, P. Papadopoulos, UC Regents Rocks Visualization “Viz” Wall Enables LCD Clusters  One PC / tile  Gigabit Ethernet  Tile Frame Applications  Large remote sensing  Volume Rendering  Seismic Interpretation Electronic Visualization Lab  Bio-Informatics  Bio-Imaging (NCMIR BioWall)

Rockstar

Copyright © 2004 F. Sacerdoti, M. Katz, G. Bruno, P. Papadopoulos, UC Regents Rockstar Cluster Collaboration between SDSC and SUN 129 Nodes: Sun V60x (Dual P4 Xeon)  Gigabit Ethernet Networking (copper)  Top500 list positions: 201, 433 Built on showroom floor of Supercomputing Conference 2003  Racked, Wired, Installed: 2 hrs total  Running apps through SGE

Copyright © 2004 F. Sacerdoti, M. Katz, G. Bruno, P. Papadopoulos, UC Regents Building of Rockstar

Copyright © 2004 F. Sacerdoti, M. Katz, G. Bruno, P. Papadopoulos, UC Regents Rockstar Topology 24-port switches Not a symmetric network  Best case - 4:1 bisection bandwidth  Worst case - 8:1  Average - 5.3:1 Linpack achieved 49% of peak Very close to percentage peak of 1st generation DataStar at SDSC

Copyright © 2004 F. Sacerdoti, M. Katz, G. Bruno, P. Papadopoulos, UC Regents Rocks Future Work High Availability: N Frontend nodes.  Not that far off (supplemental install server design)  Limited by Batch System  Frontends are long lived in practice:  Keck 2 Cluster (UCSD) uptime: 249 days, 2:56 Extreme install scaling More Rolls! Refinements

Copyright © 2004 F. Sacerdoti, M. Katz, G. Bruno, P. Papadopoulos, UC Regents Rocks mailing List  Rocks Cluster Register  Core: