Introduce yourself Presented by

Slides:



Advertisements
Similar presentations
QCloud Queensland Cloud Data Storage and Services 27Mar2012 QCloud1.
Advertisements

Windows Deployment Services WDS for Large Scale Enterprises and Small IT Shops Presented By: Ryan Drown Systems Administrator for Krannert.
1 Storage Today Victor Hatridge – CIO Nashville Electric Service (615)
Duke Atlas Tier 3 Site Doug Benjamin (Duke University)
Barracuda Backup Service Data Backup and Disaster Recovery.
© Copyright 2011 John Wiley & Sons, Inc.
Research Computing with Newton Gerald Ragghianti Nov. 12, 2010.
Chapter 13 Organizing Information System Resources MIS Department Centralization and Decentralization Outsourcing Computer Facilities and Services.
VAP What is a Virtual Application ? A virtual application is an application that has been optimized to run on virtual infrastructure. The application software.
Chapter 11: Physical Architecture Layer Design
F Run II Experiments and the Grid Amber Boehnlein Fermilab September 16, 2005.
Open Science Grid For CI-Days Internet2: Fall Member Meeting, 2007 John McGee – OSG Engagement Manager Renaissance Computing Institute.
DRAFT 1 Institutional Research Computing at WSU: A community-based approach Governance model, access policy, and acquisition strategy for consideration.
LOCAL AREA NETWORK A local area network (lan) is a communication network that interconnects a variety of data communicating devices within a small geographic.
PCGRID ‘08 Workshop, Miami, FL April 18, 2008 Preston Smith Implementing an Industrial-Strength Academic Cyberinfrastructure at Purdue University.
Ch 5. The Evolution of Analytic Processes
Ocean Observatories Initiative Common Execution Infrastructure (CEI) Overview Michael Meisinger September 29, 2009.
Storage and data services eIRG Workshop Amsterdam Dr. ir. A. Osseyran Managing director SARA
Open Science Grid For CI-Days Elizabeth City State University Jan-2008 John McGee – OSG Engagement Manager Manager, Cyberinfrastructure.
14 Aug 08DOE Review John Huth ATLAS Computing at Harvard John Huth.
Developing & Managing A Large Linux Farm – The Brookhaven Experience CHEP2004 – Interlaken September 27, 2004 Tomasz Wlodek - BNL.
CERN Physics Database Services and Plans Maria Girone, CERN-IT
Authors: Ronnie Julio Cole David
OSG Tier 3 support Marco Mambelli - OSG Tier 3 Dan Fraser - OSG Tier 3 liaison Tanya Levshina - OSG.
BNL Tier 1 Service Planning & Monitoring Bruce G. Gibbard GDB 5-6 August 2006.
Ruth Pordes November 2004TeraGrid GIG Site Review1 TeraGrid and Open Science Grid Ruth Pordes, Fermilab representing the Open Science.
VMware vSphere Configuration and Management v6
The LBNL Perceus Cluster Infrastructure Next Generation Cluster Provisioning and Management October 10, 2007 Internet2 Fall Conference Gary Jung, SCS Project.
CERN - IT Department CH-1211 Genève 23 Switzerland t High Availability Databases based on Oracle 10g RAC on Linux WLCG Tier2 Tutorials, CERN,
Comprehensive Scientific Support Of Large Scale Parallel Computation David Skinner, NERSC.
Maria Girone CERN - IT Tier0 plans and security and backup policy proposals Maria Girone, CERN IT-PSS.
Computing Issues for the ATLAS SWT2. What is SWT2? SWT2 is the U.S. ATLAS Southwestern Tier 2 Consortium UTA is lead institution, along with University.
+ Lec#1: Planning, Designing, and Operating Local Area Networks 1 st semester CT.
Northwest Indiana Computational Grid Preston Smith Rosen Center for Advanced Computing Purdue University - West Lafayette West Lafayette Calumet.
INTRODUCTION TO XSEDE. INTRODUCTION  Extreme Science and Engineering Discovery Environment (XSEDE)  “most advanced, powerful, and robust collection.
Input and Output Optimization in Linux for Appropriate Resource Allocation and Management James Avery King.
Unit 3 Virtualization.
Elastic Cyberinfrastructure for Research Computing
Chapter 12: Architecture
HTCondor Annex (There are many clouds like it, but this one is mine.)
XSEDE Value Added and Financial Economies
Designing the Physical Architecture
What is HPC? High Performance Computing (HPC)
TIM 58 Chapter 11: Physical Architecture Layer Design
Discovering Computers 2010: Living in a Digital World Chapter 14
Deploying Regional Grids Creates Interaction, Ideas, and Integration
Clouds , Grids and Clusters
Introduction to Distributed Platforms
Scaling Science Communities Lessons learned by and future plans of the Open Science Grid Frank Würthwein OSG Executive Director Professor of Physics UCSD/SDSC.
Working Group 4 Facilities and Technologies
Western Analysis Facility
Grid Computing.
NGS Oracle Service.
THE STEPS TO MANAGE THE GRID
LQCD Computing Operations
Physical Architecture Layer Design
Building a Cyberinfrastructure Culture: IT as a Partner in Research
Introduction.
Introduction to client/server architecture
Infrastructure, Data Center & Managed Services
Grid Means Business OGF-20, Manchester, May 2007
Systems Analysis and Design 5th Edition Chapter 8. Architecture Design
Your Next LIMS: SaaS or On-Premise? Presented by:
Mary Fran Yafchak Senior Program Manager, IT
Introduction to Networks
Chapter 12: Physical Architecture Layer Design
Your Facility Your Information
Resource Allocation in a Middleware for Streaming Data
Salesforce.com Salesforce.com is the world leader in on-demand customer relationship management (CRM) services Manages sales, marketing, customer service,
Michael Westwater SENIOR PLANNER – National Policy Team
Presentation transcript:

Transition in Campus CyberInfrastructure: Community Clusters, Storage and Co-Loc Introduce yourself Presented by Dwight McKay, Director of Systems Engineering ITaP Rosen Center for Advanced Computing Purdue University

Introduction Community Clusters Summarize model and issues Outline Introduction Community Clusters Summarize model and issues Perspective is that of an infrastructure builder / operator Focus on computational and storage allocation, growth strategies, funding Start with our structure and the sea change we see as we move from a centrally funded world to being a resource provider on-campus and beyond

Introduction Community Clusters Summary This is the overall structure of ITaP Introduction Community Clusters Summary

Introduction Community Clusters Summary The Rosen Center is one of five business units. We focus on the cyberinfrastructure needs of the campus and beyond Introduction Community Clusters Summary

Rosen Center for Advanced Computing This is the structure of RCAC Talk about each sub-unit Introduction Community Clusters Summary

Rosen Center for Advanced Computing This is the structure of RCAC Talk about each sub-unit Introduction Community Clusters Summary

Transition in Research Computing Support Move from the center to a service, a collaborator Our history has been to share systems, first with job scheduling, then as we transitioned out of the centrally funded realm we built clusters from old lab systems and then moved into the condo model that we call community clusters. Introduction Community Clusters Summary

Transition in Research Computing Support Change in Direction From central purchase to researcher / project purchase From central shared facility to resource or service provider From service desk to partner Move from the center to a service, a collaborator Our history has been to share systems, first with job scheduling, then as we transitioned out of the centrally funded realm we built clusters from old lab systems and then moved into the condo model that we call community clusters. Introduction Community Clusters Summary

Transition Implications Paying Customers Higher Expectations Service & Support Formal Agreements Cultural Change This transition is the driver for moving into new models of systems acquisition, resource allocation and collaboration The biggest change is that we now have customers explicitly paying for services Introduction Community Clusters Summary

Rosen Center for Advanced Computing While we are structured into specific areas, the boundaries between these areas are more fluid than this diagram suggests. The structure is more of a matrix with projects and people spanning across the reporting boundaries as needed to support our customers. Also note that we incorporate a research group as well as user support. Research User Support Infrastructure Introduction Community Clusters Summary

Rosen Center for Advanced Computing Project A typical project has connections into both computational infrastructure, accounts, queues, etc. AND high level support, consultation, code optimization, application support, etc. Larger, more complex projects often pull in larger sets of resources, such as project specific WAN links, project management, software development and custom infrastructure design and deployment. Project Introduction Community Clusters Summary

Rosen Center for Advanced Computing Our teams have people who span groups. A person reporting to Seb to manage a system in a grid project would also participate in the system team, come to our meetings, work and act like a member of our team to provide a close connection and better achieve the support needs the grid project needs. We also embed people into research teams to provide IT expertise needed to move a project along. Project Introduction Community Clusters Summary

Transition Implications New Business Models Needed HW/SW, infrastructure, support? Unpredictable Demand & Funding Planning for power/space/cooling? Non-paying “Users” How do we pay for hardware, software, services, people in this new environment? If we are not centrally funded, how do we predict the demand we will see from our customers? What about our “general” users; those who did not buy into our services? Introduction Community Clusters Summary

Custom Arrangements for Specific Projects Community Clusters Condominium Model Purchase Computation “by the node” Nodes come with bundle of services Purchase Storage “by the TeraByte” Storage comes with bundle of services Custom Arrangements for Specific Projects Introduction Community Clusters Summary

Services Community Clusters HW Installation HW Maintenance Facilities Support Network Connection Infiniband Connnection OS installation and management Disk Storage Archival Storage On-Call Support Disaster Recovery Security Introduction Community Clusters Summary

Services Node Bundle Community Clusters HW Installation HW Maintenance Facilities Support Network Connection Infiniband Connnection OS installation and management Disk Storage Archival Storage On-Call Support Disaster Recovery Security Node Bundle Introduction Community Clusters Summary

Tiered Cycle Allocation Community Clusters Tiered Cycle Allocation Owners guaranteed specific share Owners given first pick of idle cycles Owners agree to harvesting remaining idle cycles “Use the whole buffalo.” -- Brad Bird Brad Bird is the director of “The Incredibles”. Important for space/power/cooling, non-paying customers, cycles for Grid users Introduction Community Clusters Summary

Tiered Cycle Allocation Community Clusters Tiered Cycle Allocation Owner Introduction Community Clusters Summary

Tiered Cycle Allocation Community Clusters Tiered Cycle Allocation Owner Pre-empt Introduction Community Clusters Summary

Tiered Cycle Allocation Community Clusters Tiered Cycle Allocation Owner Mention Condor here. Cycle Harvesting Pre-empt Introduction Community Clusters Summary

Community Clusters Challenges Business model that spans system generations & recovers shared infrastructure costs Cluster Heterogeneity Multiple communities sharing one cluster (TeraGrid, OSG, NWICG) Other architectures & special needs? We are heading towards the end of a three year cycle on cluster building. How do we do retirement? How do we pay for interconnection infrastructure and upgrade it over time? What do we do when the particular node we use is nolonger available? What about those folks who need a shared memory system? Introduction Community Clusters Summary

Storage Community Clusters Three Primary Tiers Fast for scratch files Commodity for home directories Archival for “longer term” storage Custom Storage for Specific Projects Introduction Community Clusters Summary

Storage Challenges Community Clusters Business model that spans lifetime of data / researcher? Media Purchase vs. Space Rental? Data Retention Policy Initially we had researchers buy disk trays. But storage technology is progressing and there’s a potential danger in being stuck maintaining old storage. How long to keep something and how do we help a research take his data with him when he might have 10s to 100s of TB? Introduction Community Clusters Summary

Connectivity Challenges Community Clusters Connectivity Challenges Multiple classes of network are needed Direct Routes Data Center to Key Research WANs Data Center to Key Research Labs Introduction Community Clusters Summary

Introduction Community Clusters Summary

Four Fold Network Community Clusters Commodity Secure Research High Performance / Large Data Network Research Low Latency Introduction Community Clusters Summary

Summary Transition to resource provider / partner / service Architecture Computation -> Community Clusters Storage -> 3 Tiers Connectivity -> Direct Data Center connections to Research WAN and Lab Access Challenges Customer expectations and implications Serving up storage and other architectures “by the slice” Recovering ALL the costs, especially support Introduction Community Clusters Summary