NCCS User Forum June 15, 2010. Agenda Current System Status Fred Reitz, HPC Operations NCCS Compute Capabilities Dan Duffy, Lead Architect User Services.

Slides:



Advertisements
Similar presentations
Living with Exadata Presented by: Shaun Dewberry, OS Administrator, RDC Tom de Jongh van Arkel, Database Administrator, RDC Komaran Hansragh, Data Warehouse.
Advertisements

MUNIS Platform Migration Project WELCOME. Agenda Introductions Tyler Cloud Overview Munis New Features Questions.
QCloud Queensland Cloud Data Storage and Services 27Mar2012 QCloud1.
ASCR Data Science Centers Infrastructure Demonstration S. Canon, N. Desai, M. Ernst, K. Kleese-Van Dam, G. Shipman, B. Tierney.
National Center for Atmospheric Research John Clyne 4/27/11 4/26/20111.
NCCS User Forum June 19, Agenda Introduction Discover Updates NCCS Operations & User Services Updates Question & Answer Breakout Session: –Climate.
Near-Term NCCS & Discover Cluster Changes and Integration Plans: A Briefing for NCCS Users October 30, 2014.
LinkSCEEM-2: A computational resource for the development of Computational Sciences in the Eastern Mediterranean Mostafa Zoubi SESAME SESAME – LinkSCEEM.
Copyright 2009 FUJITSU TECHNOLOGY SOLUTIONS PRIMERGY Servers and Windows Server® 2008 R2 Benefit from an efficient, high performance and flexible platform.
Discover Cluster Upgrades: Hello Haswells and SLES11 SP3, Goodbye Westmeres February 3, 2015 NCCS Brown Bag.
NCCS User Forum September 14, Agenda – September 14, 2010 Welcome & Introduction (Phil Webster, CISTO Chief) Current System Status (Fred Reitz,
Academic and Research Technology (A&RT)
UNIVERSITY of MARYLAND GLOBAL LAND COVER FACILITY High Performance Computing in Support of Geospatial Information Discovery and Mining Joseph JaJa Institute.
Cluster computing facility for CMS simulation work at NPD-BARC Raman Sehgal.
Administration and management of Windows-based clusters Windows HPC Server 2008 Matej Ciesko HPC Consultant, PM
Hands-On Microsoft Windows Server 2008 Chapter 1 Introduction to Windows Server 2008.
Project Overview:. Longhorn Project Overview Project Program: –NSF XD Vis Purpose: –Provide remote interactive visualization and data analysis services.
The Creation of a Big Data Analysis Environment for Undergraduates in SUNY Presented by Jim Greenberg SUNY Oneonta on behalf of the SUNY wide team.
IPlant Collaborative Tools and Services Workshop iPlant Collaborative Tools and Services Workshop Collaborating with iPlant.
Towards a Comprehensive Environment for Data Analysis and Visualization Pamela Gillman, John Clyne Users Forum May 19, 2005.
SDSC RP Update TeraGrid Roundtable Reviewing Dash Unique characteristics: –A pre-production/evaluation “data-intensive” supercomputer based.
Presented by The Earth System Grid: Turning Climate Datasets into Community Resources David E. Bernholdt, ORNL on behalf of the Earth System Grid team.
ISG We build general capability Introduction to Olympus Shawn T. Brown, PhD ISG MISSION 2.0 Lead Director of Public Health Applications Pittsburgh Supercomputing.
NCCS User Forum 15 May NCCS User Forum5/15/20082 Agenda Welcome & Introduction Phil Webster NCCS Current System Status Fred Reitz, Operations Manager.
Cloud Computing in NASA Missions Dan Whorton CTO, Stinger Ghaffarian Technologies June 25, 2010 All material in RED will be updated.
Introduction to Apache OODT Yang Li Mar 9, What is OODT Object Oriented Data Technology Science data management Archiving Systems that span scientific.
NCCS NCCS User Forum 24 March NCCS Agenda Welcome & Introduction Phil Webster, CISTO Chief Current System Status Fred Reitz, Operations Manager.
Taking the Complexity out of Cluster Computing Vendor Update HPC User Forum Arend Dittmer Director Product Management HPC April,
The PROGRESS Grid Service Provider Maciej Bogdański Portals & Portlets 2003 Edinburgh, July 14th-17th.
NCCS User Forum March 20, Agenda – March 20, 2012 Welcome & Introduction (Lynn Parnell) Discover Update (Dan Duffy) Dirac DMF Growth Rate – Issue.
CRISP & SKA WP19 Status. Overview Staffing SKA Preconstruction phase Tiered Data Delivery Infrastructure Prototype deployment.
Looking Ahead: A New PSU Research Cloud Architecture Chuck Gilbert - Systems Architect and Systems Team Lead Research CI Coordinating Committee Meeting.
GEM Portal and SERVOGrid for Earthquake Science PTLIU Laboratory for Community Grids Geoffrey Fox, Marlon Pierce Computer Science, Informatics, Physics.
Tool Integration with Data and Computation Grid GWE - “Grid Wizard Enterprise”
Large Scale Parallel File System and Cluster Management ICT, CAS.
NCCS User Forum 11 December GSFC NCCS NCCS User Forum12/11/082 Agenda Welcome & Introduction Phil Webster, CISTO Chief Current System Status Fred.
NA-MIC National Alliance for Medical Image Computing UCSD: Engineering Core 2 Portal and Grid Infrastructure.
CLASS Information Management Presented at NOAATECH Conference 2006 Presented by Pat Schafer (CLASS-WV Development Lead)
IODE Ocean Data Portal - ODP  The objective of the IODE Ocean Data Portal (ODP) is to facilitate and promote the exchange and dissemination of marine.
Nanco: a large HPC cluster for RBNI (Russell Berrie Nanotechnology Institute) Anne Weill – Zrahia Technion,Computer Center October 2008.
GCRC Meeting 2004 BIRN Coordinating Center Software Development Vicky Rowley.
NCCS User Forum July 19, Agenda Introduction (Lynn Parnell, NCCS HPC Lead) Discover Update Archive Update User Services Update Data Services & Analysis.
Economic and On Demand Brain Activity Analysis on Global Grids A case study.
GSFC NCCS NCCS User Forum 25 September GSFC NCCS NCCS User Forum9/25/082 Agenda Welcome & Introduction Phil Webster, CISTO Chief Scott Wallace,
1 Accomplishments. 2 Overview of Accomplishments  Sustaining the Production Earth System Grid Serving the current needs of the climate modeling community.
Comprehensive Scientific Support Of Large Scale Parallel Computation David Skinner, NERSC.
ISG We build general capability Introduction to Olympus Shawn T. Brown, PhD ISG MISSION 2.0 Lead Director of Public Health Applications Pittsburgh Supercomputing.
CD FY09 Tactical Plan Status FY09 Tactical Plan Status Report for Neutrino Program (MINOS, MINERvA, General) Margaret Votava April 21, 2009 Tactical plan.
Welcome to the PRECIS training workshop
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”
MODIS SDST, STTG and SDDT MODIS Science Team Meeting (Land Discipline Breakout Session) July 13, 2004 Robert Wolfe Raytheon NASA GSFC Code 922.
Introduction to Hartree Centre Resources: IBM iDataPlex Cluster and Training Workstations Rob Allan Scientific Computing Department STFC Daresbury Laboratory.
Climate-SDM (1) Climate analysis use case –Described by: Marcia Branstetter Use case description –Data obtained from ESG –Using a sequence steps in analysis,
1 2.5 DISTRIBUTED DATA INTEGRATION WTF-CEOP (WGISS Test Facility for CEOP) May 2007 Yonsook Enloe (NASA/SGT) Chris Lynnes (NASA)
NCCS User Forum December 7, Agenda – December 7, 2010 Welcome & Introduction (Phil Webster, CISTO Chief) Current System Status (Fred Reitz, NCCS.
Architecture of a platform for innovation and research Erik Deumens – University of Florida SC15 – Austin – Nov 17, 2015.
Introduction to Data Analysis with R on HPC Texas Advanced Computing Center Feb
Research and Service Support Resources for EO data exploitation RSS Team, ESRIN, 23/01/2013 Requirements for a Federated Infrastructure.
High Performance Storage System (HPSS) Jason Hick Mass Storage Group HEPiX October 26-30, 2009.
 1- Definition  2- Helpdesk  3- Asset management  4- Analytics  5- Tools.
What is HPC? High Performance Computing (HPC)
Working With Azure Batch AI
LinkSCEEM-2: A computational resource for the development of Computational Sciences in the Eastern Mediterranean Mostafa Zoubi SESAME Outreach SESAME,
Provisioning 160,000 cores with HEPCloud at SC17
Appro Xtreme-X Supercomputers
Shared Research Computing Policy Advisory Committee (SRCPAC)
Data Management Components for a Research Data Archive
Welcome to (HT)Condor Week #19 (year 34 of our project)
Presentation transcript:

NCCS User Forum June 15, 2010

Agenda Current System Status Fred Reitz, HPC Operations NCCS Compute Capabilities Dan Duffy, Lead Architect User Services Updates Tyler Simon, User Services Questions and Comments Lynn Parnell/Phil Webster Welcome & Introduction Phil Webster, CISTO Chief Analysis Updates & 3D Demo Tom Maxwell, Analysis Lead Visualization Wall Demos & NCCS Facility Tours Informal Breakout Sessions NCCS User Forum, June 15, 20102

NASA Center for Climate Simulation New name for a Goddard Space Flight Center organization that has provided supercomputing resources to NASA scientists and engineers for more than 25 years. With climate and weather modeling representing the bulk of NCCS computing, the new name reflects emphasis on supporting NASA Earth science. Data-centric science: create datasets and make them available for science researchers around the world. The NCCS is continuing to expand its services to meet NASA's growing climate data needs in creation, utilization, and exploitation of Earth science model data. NCCS User Forum, June 15, 20103

Large scale HPC computing Comprehensive toolsets for job scheduling and monitoring Large capacity storage Tools to manage and protect data Data migration support Help Desk Account/Allocation support Computational science support User teleconferences Training & tutorials Interactive analysis environment Software tools for image display Easy access to data archive Specialized visualization support Internal high speed interconnects for HPC components High-bandwidth to NCCS for GSFC users Multi-gigabit network supports on-demand data transfers HPC Computing Data Archival and Stewardship Code repository for collaboration Environment for code development and test Code porting and optimization support Web based tools Code Development* Analysis & Visualization* User Services* Capability to share data & results Supports community-based development Data distribution and publishing Data Sharing and Publication Data Transfer DATA Storage & Management Global file system enables data access for full range of modeling and analysis activities * Joint effort with SIVO NCCS Data Centric Climate Simulation Environment NCCS User Forum, June 15, 20104

NCCS Compute Capacity Evolution NCCS User Forum, June 15, 20105

Highlights of the New NCCS Discover SCU7 hardware, Fall 2010 –Approximately doubles Discover capacity, to 300+ peak TFLOPS Visualization Wall in the Data Exploration Theater –Present highly detailed depictions of scientific results Data Management via Integrated Rule- Oriented Data System (iRODS) –MODIS and IPCC pilot projects Earth System Grid and other Data Portal data sharing and publishing services Advances in Data Analysis software in close collaboration with scientists NCCS User Forum, June 15, 20106

NCCS Staff Additions Maria Ealey, Computing Allocations NCCS User Forum, June 15, 20107

Meeting Logistics Questions How many for Vis Wall Demo and NCCS Facility Tours? How many for the following Informal Breakouts: –Archive data flow –Discover workflow evaluation –3D visualization –Discover nobackup plans discussion –Sysadmin confab NCCS User Forum, June 15, 20108

Agenda Current System Status Fred Reitz, HPC Operations NCCS Compute Capabilities Dan Duffy, Lead Architect User Services Updates Tyler Simon, User Services Questions and Comments Lynn Parnell/Phil Webster Welcome & Introduction Phil Webster, CISTO Chief Analysis Updates & 3D Demo Tom Maxwell, Analysis Lead Visualization Wall Demos & NCCS Facility Tours Informal Breakout Sessions NCCS User Forum, June 15, 20109

Key Accomplishments General GLOPAC mission support Data Exploration Theater Discover/Analysis Environment Returned SCU5 to normal queue structure Additional storage Dataportal Upgraded GPFS Upgraded OS to SLES 11 Established two new FTP sites – g5_cld, oceanwinds DMF Upgraded OS to SLES 10 SP3 Removed 64 processors from palm/dirac NCCS User Forum, June 15,

Discover Total CPU Consumption Past 12 Months (CPU Hours) NCCS User Forum, June 15,

Discover CPU Consumption by Group Past 12 Months NCCS User Forum, June 15,

Discover CPU Consumption by Queue Past 12 Months NCCS User Forum, June 15,

Discover Job Analysis – May 2010 NCCS User Forum, June 15,

Discover Job Analysis – May 2010 NCCS User Forum, June 15,

Discover Availability – Past 12 Months Scheduled Maintenance: March-May 22 April – 1 hour Network switch maintenance 12 May – 12.5 hours InfiniBand, disk controller firmware Unscheduled Outages: March-May 21 April – 1 hour 15 minutes GPFS hang NCCS User Forum, June 15,

NCCS Network Availability Past 12 Months NCCS User Forum, June 15,

Future Enhancements Data Exploration Theater performance, functionality Discover Cluster –PBS v10 –SLES11 –GPUs Data Portal –iRODS –Additional storage –New database servers Mass Storage –Additional storage –Platform upgrade NCCS User Forum, June 15,

Agenda Current System Status Fred Reitz, HPC Operations NCCS Compute Capabilities Dan Duffy, Lead Architect User Services Updates Tyler Simon, User Services Questions and Comments Lynn Parnell/Phil Webster Welcome & Introduction Phil Webster, CISTO Chief Analysis Updates & 3D Demo Tom Maxwell, Analysis Lead Visualization Wall Demos & NCCS Facility Tours Informal Breakout Sessions NCCS User Forum, June 15,

SCU7 Discover Cluster Augmentation SCU7 will augment Discover compute nodes with Intel Xeon Westmere processors within the existing SCU5 and SCU6 InfiniBand communication fabric. –Dual-socket, hex-core processors. –12 MB L3 cache, 50% larger than Nehalem’s. –24 GB or RAM. –Quad-Data Rate (QDR) Infiniband Initial performance of the GEOS-5 Cubed- Sphere benchmark showed a slight improvement on Westmere processors compared to Nehalem. –Test compared use of 4 cores per socket on both Nehalem and Westmere. Expecting equipment delivery July 2010 Xeon Processor SpeedCores/ Socke t L3 Cache Size Nehalem2.8 GHz48 MB Westmere2.8 GHz612 MB NCCS User Forum, June 15, Intel Xeon 5600 processor series: “Westmere” Discover’s SCU7 will augment the compute capacity available for high resolution GEOS-5 Cubed Sphere runs and for GISS and GMAO simulations in support of the IPCC 5 th Reassessment.

Extending the IB Fabric to Include SCU7 288-port DDR Qlogic IB Full Non-Blocking 288-port DDR Qlogic IB Full Non-Blocking 258 Compute Nodes 2,064 Cores 258 Compute Nodes 2,064 Cores Top Level IB Switch 288-port DDR Qlogic IB (GFE) Top Level IB Switch 288-port DDR Qlogic IB (GFE) 24 Uplinks 258 Compute Nodes 2,064 Cores 258 Compute Nodes 2,064 Cores 24 Uplinks 258 Compute Nodes 2,064 Cores 258 Compute Nodes 2,064 Cores 24 Uplinks 258 Compute Nodes 2,064 Cores 258 Compute Nodes 2,064 Cores 24 Uplinks 20 I/O Nodes 24 Service Nodes 2 Management Nodes 2 Subnet Manager Nodes 240 Compute Nodes 2,880 Cores 240 Compute Nodes 2,880 Cores QDR 2-to-1 Blocking QDR 2-to-1 Blocking 288-port DDR Qlogic IB Full Non-Blocking 288-port DDR Qlogic IB Full Non-Blocking 288-port DDR Qlogic IB Full Non-Blocking 288-port DDR Qlogic IB Full Non-Blocking 288-port DDR Qlogic IB Full Non-Blocking 288-port DDR Qlogic IB Full Non-Blocking 24 Uplinks SCU7 5 Additional Units 14,400 Cores 161 TF Peak NCCS User Forum, June 15,

NCCS Peak Computing (TF) Over Time NCCS User Forum, June 15,

NCCS Visualization Capability The NCCS High Performance Visualization Capability was installed and made operational a very short time following the early March equipment delivery, and includes: High Performance 3x5 Hyperwall (6’ x 17’) Dell servers with Nvidia FX1700 GPUs. Samsung 46-inch, very small bezel televisions, 720p. Two fixed and three mobile Scientific Analysis Workstations Dell workstations with Nvidia FX4800 GPUs. High-definition 46-inch, 1080p Samsung monitor attached to the Nvidia GPU, will be 3D capable. NCCS and SIVO’s Scientific Visualization Studio staff are providing tools and assistance to help scientists exploit the Hyperwall. Mobile workstations can be deployed for periodic use in scientists’ offices or meeting spaces. NCCS User Forum, June 15, The NCCS High Performance Visualization Capability will expedite highly detailed depictions of scientific results to facilitate the development of new insights and discoveries.

Upgrades to the Archive (Dirac) Dirac is currently on a system that is 5 years old Upgrades –New DMF server to replace the Bx2 –Parallel Data Movers for Tape –Two Login Nodes for interactive access to the archive data –Two NFS Nodes for improved NFS access to the archive data from Discover –All systems are commodity Xeon based –Upgrade to the primary disk cache to 600 TB RAW –Upgrade to the disk cache for the DMF databases for improved serviceability and access to data What does this mean to the users? –A much higher performing archive –Outages for moving file systems and databases between systems (please be patient) Time Line –Upgrades to be complete by the end of August NCCS User Forum, June 15,

Agenda Current System Status Fred Reitz, HPC Operations NCCS Compute Capabilities Dan Duffy, Lead Architect User Services Updates Tyler Simon, User Services Questions and Comments Lynn Parnell/Phil Webster Welcome & Introduction Phil Webster, CISTO Chief Analysis Updates & 3D Demo Tom Maxwell, Analysis Lead Visualization Wall Demos & NCCS Facility Tours Informal Breakout Sessions NCCS User Forum, June 15,

Ticketing System Upgrade  NCCS Users will be able to login (via the web) and create new tickets, add a new description or add an attachment to an existing open ticket.  Users will be able to search the built-in knowledge base for solutions to common problems.  Follow current system issues in real time by subscribing to global tickets. NCCS User Forum, June 15,

Discover Hardware/Software Migrations  Deadline to migrate off of ScaliMPI nodes is July 6.  We will also be moving away from the Visual nodes and Queue at a later date TBD.  Testing new versions of Intel MPI 4.0 with debugging tools. NCCS User Forum, June 15,

iRODS NCCS User Forum, June 15,

iRODS NCCS User Forum, June 15,  Analyze the potential for iRODS to federate climate simulation data and observational data  Provide one-stop shopping to remote and local data while providing data management services).  Metadata is catalogued by iRODS and that existing user interfaces let users manage the data and metadata (e.g., add, search, extract, annotate).

iRODS NCCS User Forum, June 15,

Agenda Current System Status Fred Reitz, HPC Operations NCCS Compute Capabilities Dan Duffy, Lead Architect User Services Updates Tyler Simon, User Services Questions and Comments Lynn Parnell/Phil Webster Welcome & Introduction Phil Webster, CISTO Chief Analysis Updates & 3D Demo Tom Maxwell, Analysis Lead Visualization Wall Demos & NCCS Facility Tours Informal Breakout Sessions NCCS User Forum, June 15,

Earth System Grid Data Node Update ESG Data Node software (2.4) received from PCMDI. –Installed and tested on the NCCS Dataportal. –Successfully published to PCMDI ESG Gateway. Developing plans for publishing MERRA data. –Publishing virtual files not currently feasible. –Will copy and publish monthly means. Will publish GISS & GMAO IPCC AR5 contributions. –Scheduled date for publication of initial data: August 2010 NCCS User Forum, June 15,

DV3D Interactive 3D visualization of simulation data –Tailored for climate scientists –Interface to NCCS/ESG Analysis Products Server Python app built on MayaVi / VTK –Runs inside Grads –Access all Grads variables Developed at NCCS –Hyper wall demo to follow 33 NCCS User Forum, June 15, 2010

Analysis Products Server (Under Development) S erver-side Analysis Services executed on Dali Python client for GrADS/DV3D DET: “Gateway to Dali” –Phase 1: Access saved simulation data –Phase 2: Access running simulation. Service APIs: –Live Access Server (LAS) request protocol –IRODS Planned Services –Subsetting, regridding/rescaling, reformatting, aggregation, etc. –Differencing, averaging (e.g. zonal means) –Data transforms ( wavelet, FFT, etc. ) NCCS User Forum, June 15,

Live Access Server ESG Analysis module –ESG LAS not yet available Three Tiered architecture –Interface Server Installed on dp1 –Product server Installed on dp1 –Back End Server (BES) Installed on dp7 (future: dali) Current BES uses Ferret –Potential BE Services: Grads, CDAT, NCO, IRODS NCCS User Forum, June 15,

IRODS NCCS User Forum, June 15, Integrated Rule Oriented Data System Rule Engine applies user-defined Rules and Services Nodes to be installed on Dali and Data Portal

Agenda Current System Status Fred Reitz, HPC Operations NCCS Compute Capabilities Dan Duffy, Lead Architect User Services Updates Tyler Simon, User Services Questions and Comments Lynn Parnell/Phil Webster Welcome & Introduction Phil Webster, CISTO Chief Analysis Updates & 3D Demo Tom Maxwell, Analysis Lead Visualization Wall Demos & NCCS Facility Tours Informal Breakout Sessions NCCS User Forum, June 15,

Important Contact Information NCCS Support: (301) NCCS User Forum, June 15,

Agenda Current System Status Fred Reitz, HPC Operations NCCS Compute Capabilities Dan Duffy, Lead Architect User Services Updates Tyler Simon, User Services Questions and Comments Lynn Parnell/Phil Webster Welcome & Introduction Phil Webster, CISTO Chief Analysis Updates & 3D Demo Tom Maxwell, Analysis Lead Visualization Wall Demos & NCCS Facility Tours Informal Breakout Sessions NCCS User Forum, June 15,

Vis Wall Demo and NCCS Facility Tours No Food or Drink, please Vis Wall in S214/Data Exploration Theater –Introduction (Phil Webster) –3D Demonstrations (Tom Maxwell) NCCS Facility Tours –Begin at the bottom of front stairs Informal Breakouts to follow in E210, S216, and staff offices NCCS User Forum, June 15,

Agenda Current System Status Fred Reitz, HPC Operations NCCS Compute Capabilities Dan Duffy, Lead Architect User Services Updates Tyler Simon, User Services Questions and Comments Lynn Parnell/Phil Webster Welcome & Introduction Phil Webster, CISTO Chief Analysis Updates & 3D Demo Tom Maxwell, Analysis Lead Visualization Wall Demos & NCCS Facility Tours Informal Breakout Sessions NCCS User Forum, June 15,