© 2007 UC Regents1 Rocks – Present and Future The State of Things Open Source Grids and Clusters Conference Philip Papadopoulos, Greg Bruno Mason Katz,

Slides:



Advertisements
Similar presentations
Ljubomir Ivaniš CPU d.o.o.
Advertisements

1 Applications Virtualization in VPC Nadya Williams UCSD.
© UC Regents 2010 Extending Rocks Clusters into Amazon EC2 Using Condor Philip Papadopoulos, Ph.D University of California, San Diego San Diego Supercomputer.
Notes to the presenter. I would like to thank Jim Waldo, Jon Bostrom, and Dennis Govoni. They helped me put this presentation together for the field.
City University London
Microsoft Load Balancing and Clustering. Outline Introduction Load balancing Clustering.
VMware vCenter Server Module 4.
INTRODUCTION TO CLOUD COMPUTING Cs 595 Lecture 5 2/11/2015.
CLOUD COMPUTING. A general term for anything that involves delivering hosted services over the Internet. And Cloud is referred to the hardware and software.
Windows Azure Networking & Active Directory Nasir (Muhammad Nasiruddin) Developer Evangelist - Azure Microsoft Corporation
Cloud and Virtualization Panel Philip Papadopoulos UC San Diego.
Assessment of Core Services provided to USLHC by OSG.
Rocks Clusters SUN HPC Consortium November 2004 Federico D. Sacerdoti Advanced CyberInfrastructure Group San Diego Supercomputer Center.
Rocks cluster : a cluster oriented linux distribution or how to install a computer cluster in a day.
Linux Operations and Administration

Introduction To Windows Azure Cloud
INSTALLING MICROSOFT EXCHANGE SERVER 2003 CLUSTERS AND FRONT-END AND BACK ‑ END SERVERS Chapter 4.
Cloud Computing 1. Outline  Introduction  Evolution  Cloud architecture  Map reduce operation  Platform 2.
So, Jung-ki Distributed Computing System LAB School of Computer Science and Engineering Seoul National University Implementation of Package Management.
XSEDE14 Reproducibility Workshop: Reproducibility in Large Scale Computing – Where do we stand Mark R. Fahey, NICS Robert McLay, TACC XSEDE14 - Reproducibility.
Rocks ‘n’ Rolls An Introduction to Programming Clusters using Rocks © 2008 UC Regents Anoop Rajendra.
1.  PRAGMA Grid test-bed : Shares clusters which managed by multiple sites Realizes a large-scale computational environment. › Expects as a platform.
การติดตั้งและทดสอบการทำคลัสเต อร์เสมือนบน Xen, ROCKS, และไท ยกริด Roll Implementation of Virtualization Clusters based on Xen, ROCKS, and ThaiGrid Roll.
Presented by: Sanketh Beerabbi University of Central Florida COP Cloud Computing.
Windows Azure Conference 2014 Deploy your Java workloads on Windows Azure.
ArcGIS Server for Administrators
Comparison of Distributed Operating Systems. Systems Discussed ◦Plan 9 ◦AgentOS ◦Clouds ◦E1 ◦MOSIX.
Tool Integration with Data and Computation Grid GWE - “Grid Wizard Enterprise”
Server Virtualization
© 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.
Marc Fiuczynski Princeton University Marco Yuen University of Victoria PlanetLab & Clusters.
Virtual Workspaces Kate Keahey Argonne National Laboratory.
NA-MIC National Alliance for Medical Image Computing UCSD: Engineering Core 2 Portal and Grid Infrastructure.
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,
Cluster Software Overview
A. Frank - P. Weisberg Operating Systems Structure of Operating Systems.
Security Vulnerabilities in A Virtual Environment
7. Grid Computing Systems and Resource Management
Virtualization Technology and Microsoft Virtual PC 2007 YOU ARE WELCOME By : Osama Tamimi.
Scalability Requirements and Implementation Options.
TACTIC | Workflow: Project Management OSS on Microsoft Azure Helps Enterprises to Create Streamline, Manage, and Track Digital Content MICROSOFT AZURE.
Alessandro Cardoso, Microsoft MVP Creating your own “Private Cloud” with Windows 10 Hyper- V WIN443.
GraDS MacroGrid Carl Kesselman USC/Information Sciences Institute.
Tool Integration with Data and Computation Grid “Grid Wizard 2”
Microsoft Azure and DataStax: Start Anywhere and Scale to Any Size in the Cloud, On- Premises, or Both with a Leading Distributed Database MICROSOFT AZURE.
Windows Azure poDRw_Xi3Aw.
Cloud Computing – UNIT - II. VIRTUALIZATION Virtualization Hiding the reality The mantra of smart computing is to intelligently hide the reality Binary->
Predrag Buncic (CERN/PH-SFT) Software Packaging: Can Virtualization help?
T3g software services Outline of the T3g Components R. Yoshida (ANL)
Virtual Machines Module 2. Objectives Define virtual machine Define common terminology Identify advantages and disadvantages Determine what software is.
Virtualization Assessment. Strategy for web hosting Reduce costs by consolidating services onto the fewest number of physical machines
Nara Institute of Science and Technology, Nara Prefecture, Japan CONFIGURATION AND DEPLOYMENT OF A SCALABLE VIRTUAL MACHINE CLUSTER FOR MOLECULAR DOCKING.
Building on virtualization capabilities for ExTENCI Carol Song and Preston Smith Rosen Center for Advanced Computing Purdue University ExTENCI Kickoff.
vSphere 6 Foundations Exam Training
INF526: Secure Systems Administration Composition of Systems And Security Domains Prof. Clifford Neuman Lecture 3 3 June 2016 OHE100C.
Amazon Web Services. Amazon Web Services (AWS) - robust, scalable and affordable infrastructure for cloud computing. This session is about:
June 28, 2016 Cluster Management for Non-XSEDE Systems Barbara Hallock, Eric Coulter, Sudhakar Pamidighantam.
Univa Grid Engine Makes Work Management Automatic and Efficient, Accelerates Deployment of Cloud Services with Power of Microsoft Azure MICROSOFT AZURE.
A Cloudy Future Panel at CCGSC ‘08
Network Operating System Lab
GWE Core Grid Wizard Enterprise (
Cloud Computing Dr. Sharad Saxena.
Scalable SoftNAS Cloud Protects Customers’ Mission-Critical Data in the Cloud with a Highly Available, Flexible Solution for Microsoft Azure MICROSOFT.
DeFacto Planning on the Powerful Microsoft Azure Platform Puts the Power of Intelligent and Timely Planning at Any Business Manager’s Fingertips Partner.
Data Security for Microsoft Azure
Abiquo’s Hybrid Cloud Management Solution Helps Enterprises Maximise the Full Potential of the Microsoft Azure Platform MICROSOFT AZURE ISV PROFILE: ABIQUO.
Chapter 7 –Implementation Issues
Last.Backend is a Continuous Delivery Platform for Developers and Dev Teams, Allowing Them to Manage and Deploy Applications Easier and Faster MICROSOFT.
Presentation transcript:

© 2007 UC Regents1 Rocks – Present and Future The State of Things Open Source Grids and Clusters Conference Philip Papadopoulos, Greg Bruno Mason Katz, Anoop Rajendra

Personal Thanks  Thanks to everyone who is here.  Without users, we have no reason to build software  Special Thanks to Steve Tuecke. This conference/workshop would not have happened without him.  Many thanks Laura Kesselman-Jones. She has made sure that everything from registration to AV has worked © 2007 UC Regents2

YOUR role  Ask questions! This is a workshop. We don’t want to be “Send-only Mode”  Take advantage of the “Grill the Guru’s” session.  Come talk one-on-one with Rocks, Globus, and SGE developers to solve your problems © 2007 UC Regents3

4 The Goal of Rocks Enable domain scientists (non-specialists) to trivially build clusters of all functions and sizes (Put in DVDs and Start Computing) Rocks easily handles more than just MPI /HPC Clusters

© 2007 UC Regents5 Our Original Goal: Support the Traditional “Beowulf” Frontend Node Public Ethernet Private Ethernet Network Application Network (Optional) Node Power Distribution (Net addressable units as option)

© 2007 UC Regents6 The Modern “Cluster” Architecture is more Interesting

© 2007 UC Regents7 And Rocks Supports other types Clusters  Cluster of GPUs  OpenGL machine  Not an MPI machine  Massive Pixel Walls  60 MegaPixels  Full rate HDTV  Software  SAGE  DMX  Chromium

Rocks Today  Open-source Active development for 7+ years  Demonstrated Scalability from 2 – 1000 nodes. Several 1000 Clusters Worldwide  ~1800 Users on our discussion list  About 450 Messages/Month  Commercial support available, in addition.  Redhat Enterprise 4/5 compatible.  Core to many NSF/NIH-funded grants  Funding from NSF for core through 2010 © 2007 UC Regents8

Big Clusters, Little Clusters, and Everything in Between © 2007 UC Regents9

What Rocks Solves  You are building a [compute, web, database, visualization, ??] cluster  What software should I select ?  How should it be configured?  How long will it be before I start useful work ?  I like your cluster configuration, but I have a different HW vendor, how do I re-create what you have already done?  I’d rather drink coffee than configure software © 2007 UC Regents10

Rolls: Our Extension Mechanism  Rolls define the “macro” components of your cluster  Contain: Packages AND how to Automatically Configure/Localize  Can and are built by others  Many examples:  Grid (Globus), SGE  Condor, Torque, Moab  Java, Area51, Intel Compiler, Viz,  … (+Several Commercial ) © 2007 UC Regents11

Rocks Base HPC Sched uler GridBIRN Project-Specific Extension Rolls Define the Complete Stack

Rocks Version 5.0 (V)  Released 4/30/08.  Support RHEL/CentOS 5.1  Xen-based virtualization  Define, Start, Stop, Move Virtual Machines Identical description mechanism for real/virtual systems  Low-Level, but Important features  Programmatic partitioning (“infinite customization”)  Flash BIOS via PXE  Many SW Version Updates © 2007 UC Regents13

Virtualization is Having Positive and Widespread Impact, but beware  The VM Image is  One big bucket of Bits  Not Replicable if you don’t know how it was built  Not easily extensible if you don’t how it was built  Giving complete OS Power to the User runs the following Risks  Turn Every Scientist into a System Administrator  Unpatched/Vulnerable OS installations, on “well- cared” for machines © 2007 UC Regents14

Predictions  In 3 Years, Many Users Will Run Their HPC/Parallel Applications Exclusively inside of VMs  Irrespective of where the hardware is located (local, utility, etc)  In 5 years, Huge Computing Facilities (Like Teragrid) will be Mandated to Host User-defined VMs as Standard Procedure  All Performance Inadequacies Will be Deemed “Acceptable”  Rigorous Software Integration will far outweigh hardware concerns  (We’re working to make Rocks More Adaptable) © 2007 UC Regents15

© 2007 UC Regents16