IPPS 981 Berkeley FY98 Resource Working Group David E. Culler Computer Science Division U.C. Berkeley

Slides:



Advertisements
Similar presentations
Distributed Data Processing
Advertisements

Ed Duguid with subject: MACE Cloud
Presented by: Yash Gurung, ICFAI UNIVERSITY.Sikkim BUILDING of 3 R'sCLUSTER PARALLEL COMPUTER.
Unique Opportunities in Experimental Computer Systems Research - the Berkeley Testbeds David Culler U.C. Berkeley Grad.
Understanding Application Scaling NAS Parallel Benchmarks 2.2 on NOW and SGI Origin 2000 Frederick Wong, Rich Martin, Remzi Arpaci-Dusseau, David Wu, and.
NPACI Panel on Clusters David E. Culler Computer Science Division University of California, Berkeley
Microsoft Clustering Sean Roberts, Jean Pierre SLAC.
IPPS 981 What’s So Different about Cluster Architectures? David E. Culler Computer Science Division U.C. Berkeley
Energy Efficient Prefetching – from models to Implementation 6/19/ Adam Manzanares and Xiao Qin Department of Computer Science and Software Engineering.
Energy Efficient Prefetching with Buffer Disks for Cluster File Systems 6/19/ Adam Manzanares and Xiao Qin Department of Computer Science and Software.
NOW 1 Berkeley NOW Project David E. Culler Sun Visit May 1, 1998.
6/28/98SPAA/PODC1 High-Performance Clusters part 2: Generality David E. Culler Computer Science Division U.C. Berkeley PODC/SPAA Tutorial Sunday, June.
MS 9/19/97 implicit coord 1 Implicit Coordination in Clusters David E. Culler Andrea Arpaci-Dusseau Computer Science Division U.C. Berkeley.
TITAN: A Next-Generation Infrastructure for Integrating and Communication David E. Culler Computer Science Division U.C. Berkeley NSF Research Infrastructure.
Lecture 1: Introduction CS170 Spring 2015 Chapter 1, the text book. T. Yang.
70-290: MCSE Guide to Managing a Microsoft Windows Server 2003 Environment, Enhanced Chapter 8: Implementing and Managing Printers.
Chapter 8: Network Operating Systems and Windows Server 2003-Based Networking Network+ Guide to Networks Third Edition.
NPACI: National Partnership for Advanced Computational Infrastructure August 17-21, 1998 NPACI Parallel Computing Institute 1 Cluster Archtectures and.
Microsoft Load Balancing and Clustering. Outline Introduction Load balancing Clustering.
VMware vCenter Server Module 4.
CH 13 Server and Network Monitoring. Hands-On Microsoft Windows Server Objectives Understand the importance of server monitoring Monitor server.
Common Services in a network Server : provide services Type of Services (= type of servers) –file servers –print servers –application servers –domain servers.
Introduction to Networks Networking Concepts IST-200 VWCC 1.
Windows Server MIS 424 Professor Sandvig. Overview Role of servers Performance Requirements Server Hardware Software Windows Server IIS.
1 The Virtual Reality Virtualization both inside and outside of the cloud Mike Furgal Director – Managed Database Services BravePoint.
CLUSTER COMPUTING Prepared by: Kalpesh Sindha (ITSNS)
Abstract Load balancing in the cloud computing environment has an important impact on the performance. Good load balancing makes cloud computing more.
1 In Summary Need more computing power Improve the operating speed of processors & other components constrained by the speed of light, thermodynamic laws,
CERN IT Department CH-1211 Genève 23 Switzerland t Virtualization with Windows at CERN Juraj Sucik, Emmanuel Ormancey Internet Services Group.

Module 13: Configuring Availability of Network Resources and Content.
COEN 252 Computer Forensics
Module 12: Designing High Availability in Windows Server ® 2008.
Introduction and Overview Questions answered in this lecture: What is an operating system? How have operating systems evolved? Why study operating systems?
INSTALLING MICROSOFT EXCHANGE SERVER 2003 CLUSTERS AND FRONT-END AND BACK ‑ END SERVERS Chapter 4.
University of Illinois at Urbana-Champaign NCSA Supercluster Administration NT Cluster Group Computing and Communications Division NCSA Avneesh Pant
MSc. Miriel Martín Mesa, DIC, UCLV. The idea Installing a High Performance Cluster in the UCLV, using professional servers with open source operating.
CLUSTER COMPUTING STIMI K.O. ROLL NO:53 MCA B-5. INTRODUCTION  A computer cluster is a group of tightly coupled computers that work together closely.
March 3rd, 2006 Chen Peng, Lilly System Biology1 Cluster and SGE.
The Red Storm High Performance Computer March 19, 2008 Sue Kelly Sandia National Laboratories Abstract: Sandia National.
Designing and Deploying a Scalable EPM Solution Ken Toole Platform Test Manager MS Project Microsoft.
UNITED STATES. Understanding NDS for Directory- Enabled Solutions Ed Shropshire, NDS Developer Program Manager Novell, Inc.
Batch Scheduling at LeSC with Sun Grid Engine David McBride Systems Programmer London e-Science Centre Department of Computing, Imperial College.
Designing a Scalable Enterprise Project Management Architecture Ken Toole Platform Test Manager MS Project Microsoft Corporation.
RAL Site Report Andrew Sansum e-Science Centre, CCLRC-RAL HEPiX May 2004.
1 Week #10Business Continuity Backing Up Data Configuring Shadow Copies Providing Server and Service Availability.
PARALLEL COMPUTING overview What is Parallel Computing? Traditionally, software has been written for serial computation: To be run on a single computer.
CLUSTER COMPUTING TECHNOLOGY BY-1.SACHIN YADAV 2.MADHAV SHINDE SECTION-3.
Wellcome Trust Sanger Institute Informatics Systems Group Ensembl Compute Grid issues James Cuff Informatics Systems Group Wellcome Trust Sanger Institute.
Server Performance, Scaling, Reliability and Configuration Norman White.
Test Results of the EuroStore Mass Storage System Ingo Augustin CERNIT-PDP/DM Padova.
IDE disk servers at CERN Helge Meinhard / CERN-IT CERN OpenLab workshop 17 March 2003.
Millennium Executive Committee Meeting David E. Culler Computer Science Division
COMP381 by M. Hamdi 1 Clusters: Networks of WS/PC.
SimMillennium Systems Requirements and Challenges David E. Culler Computer Science Division U.C. Berkeley NSF Site Visit March 2, 1998.
HNC COMPUTING - Network Concepts 1 Network Concepts Network Concepts Network Operating Systems Network Operating Systems.
10/18/01Linux Reconstruction Farms at Fermilab 1 Steven C. Timm--Fermilab.
Condor on Dedicated Clusters Peter Couvares and Derek Wright Computer Sciences Department University of Wisconsin-Madison
Networking Week #10 OBJECTIVES Chapter #6 Questions Review Chapter #8.
Network Attached Storage Overview
Berkeley Cluster Projects
Debunking the Top 10 Myths of Small Business Server: Using Windows SBS in Larger Environments Abstract: This session will debunk some of the common myths.
IBM Pervasive Computing Visit June 9, 1997
Results of Prior NSF RI Grant: TITAN
NCSA Supercluster Administration
Design Unit 26 Design a small or home office network
Web Server Administration
SAP R/3 Installation on WIN NT-ORACLE
Distributing META-pipe on ELIXIR compute resources
Chapter-1 Computer is an advanced electronic device that takes raw data as an input from the user and processes it under the control of a set of instructions.
Presentation transcript:

IPPS 981 Berkeley FY98 Resource Working Group David E. Culler Computer Science Division U.C. Berkeley

IPPS 982 Disclaimer Still a research group, not a computer center Project transition phase 0.25 Staff FTE (- family leave) till now Finally got jobs through UCB Interviewing NOW!

IPPS 983 NPACI Users Currently 43 official NPACI users from 15 sites –still modest number of hours –front-end time is free –only account for GLUnix usage Still have many unofficial external users –about 50 non-CS at UCB and 25 external »some have both kinds of accounts –attempted to run CS267 through NPACI to debug “partnership” »process too slow, difficult to use other resources –attempted to move external NOW users to NPACI »not ready for the rush So far mostly conventional MPI users –not “systems” users yet

IPPS 984 Available Resources

IPPS 985 Partitions Initially a 10-node “default” cluster to absorb sequential load production parallel cluster –19 => 32 nodes grappling with NOW research NPACI usage

IPPS 986 Usage

IPPS 987 Hardware Development Sun 450 SMP front-end for NPACI users –now.npaci.edu –starting point for next years NPACI CLUMPS New slice of a clustered file server –Sun 450 SMP (4 GB, 4 proc) –with 500 GB fiberchannel attached drive –tape stacker for backup Testing gigabit ethernet backbone

IPPS 988 Cluster of SMPs (CLUMPS) Four Sun E5000s –8 processors –3 Myricom NICs Multiprocessor, Multi- NIC, Multi-Protocol

IPPS 989 Pleiades Information Servers Basic Storage Unit: – Ultra 2, 300 GB raid, 800 GB tape stacker, ATM –scalable backup/restore Dedicated Info Servers –web, –security, –mail, … VLANs project into dept.

IPPS 9810 Pleiades Basic Hardware Configuration Sun proc 4 GB 126 GB Sun proc 2 GB FC hub FC hub ATM... Tomorrow Bay Router Pleiades NPACI CLUMPS

IPPS 9811 Tool Development GLUnix availability enhanced through GLUguard MPI over AM protocol development Virtual Network support Implicit Coscheduling Split-C environment Performance Analysis Tools available for download

IPPS 9812 Process 3 Automatic Mgmt of Virtual Networks Collection of Endpoints form a Virtual Network Direct, protected hardware access performance General purpose Process n Process 2 Process 1 *** Host Memory Processor NIC Mem Network Interface P

IPPS 9813 MPI over AM Network ° ° ° Machine Architecture AM ADI MPI CH App Machine Architecture AM ADI MPI CH App Machine Architecture AM ADI MPI CH App

IPPS 9814 Implicit co-scheduling Obtain coordinated without explicit subsystem interaction, only the events in the program –very easy to build –potentially very robust to component failures –inherently “service on-demand” –scalable Local service component can evolve. A LS A GS A LS GS A LS A GS LS A GS

IPPS 9815 Performance Analysis Tools NPB LU-A

IPPS 9816 Tools (cont) 8-fold reduction in miss rate from 4 to 8 proc

IPPS 9817 Imports and Exports NPACI file configuration –will be finished with new server TCP wrappers –very valuable for NOW NPACI queueing SSH and Kerberos environment –shared K5 domain

IPPS 9818 FY99 Budget Summary Faculty Time1.75 mo 20.4 K Post-doc Res.1 FT 79.7 K Support Staff2.2 FT291 K Travel 19.5K 4 x 4 Clumps358 K S&E 51.6K Direct Cost Total746.1 K Growth: Scale CLUMPS to 32, 48, or 64 proc.

IPPS 9819 Millennium PC Clumps Initial phase of 6M$ Intel Grant Inexpensive, easy to manage Cluster NOW environment moving to NT Replicated in many departments Prototype for 400 proc PC cluster