Mark%20L.%20Green,%20Ph.D.

1
Center for Computational Research and New York
State Grid-Cyberinstitute

• Mark L. Green, Ph.D.
• Head of Cyberinfrastructure

2
Outline

• Primary Centers
• Enabling Technologies
• Modern Data Center Trends
• Enabling Services and Applications
• Enabling the next generation of discovery

3
Outline

• Primary Centers
• Enabling Technologies
• Modern Data Center Trends
• Enabling Services and Applications
• Enabling the next generation of discovery

4
State University of New York (SUNY)

• New York State includes a variety of private and
  public institutions, including the State
  University of New York (SUNY), which is comprised
  of
• 64 campuses,
• has an enrollment of nearly 414,000 students,
• is served by more than 30,000 faculty, and
• has annual sponsored research expenditures of
  approximately 850 million.

5
Center for Computational Research

• Leading Academic Supercomputing Site
• Enable Research at UB and in WNY
• 90 Research Groups/25 Departments
• Industrial Outreach/Technology Transfer to WNY
• Funding
• NSF, NIH, NIMA, EPA, Keck, Sloan,
  NYS, UB, SUNY
• IBM, SGI, Sun, HP, Dell, Nortel, Myricom
• Total Aggregate Computing Capacity
• Current - 22 Teraflops
• 220 Terabytes of On-line Storage
• 250 Terabytes of tape storage

6
CCR Compute/Storage Resources (22 Tflops)

• Dell Linux Cluster (10TF)
• 1600 P4 Processors (3.2 GHz)
• Myrinet Interconnect
• 2000 GB RAM
• 30 TB SAN 60 TB Local Disk
• Dell Linux Cluster (2.9TF)
• 600 P4 Processors (2.4 GHz)
• Myrinet Interconnect
• 600 GB RAM 40 TB Disk
• Dell Linux Cluster (6TF)
• 4036 Processors (PIII 1.2 GHz)
• 2TB RAM
• 16 TB SAN 160TB Local Disk
• IBM BladeCenter (3TF)
• 532 P4 Processors (2.8 GHz)
• 5TB SAN

• SGI Altix3700 (0.4TF)
• 64 Processors (1.3GHz ITF2)
• 256 GB RAM
• 2.5 TB Disk
• BioACE (bioinformatics)
• Sun V880 (3), Sun 6800
• Sun 280R (2)
• Intel P4 Servers
• Sun 3960 7 TB Disk Storage
• HP/Compaq SAN
• 75 TB Disk 190 TB Tape
• 64 Alpha Processors
• 32 GB RAM 400 GB Disk

7
New York State Center of Excellence in
Bioinformatics Life Sciences

• The 1B New York State Centers of Excellence
  program includes
• the Center of Excellence in Bioinformatics Life
  Sciences at SUNY-Buffalo,
• Syracuses Center for Excellence in Environmental
  and Energy Systems,
• Rochesters Center for Excellence in Photonics
  and Microsystems,
• Albanys Center for Excellence in
  Nanoelectronics, and
• Stony Brooks Center for Excellence in Wireless
  Information Technology.

8
Buffalo Niagara Medical Campus (BNMC)

• The BNMC has developed a unique partnership in
  which the
• University at Buffalo,
• Roswell Park Cancer Institute,
• Olmsted Center for the Visually Impaired,
• Kaleida Health,
• Hauptman-Woodward Medical Research Institute,
• Buffalo Medical Group Foundation and
• Buffalo Hearing Speech Center
• are working in collaboration with the surrounding
  neighborhoods and local government to cultivate a
  world-class medical campus in downtown Buffalo.

9
(No Transcript)
10
New York State Grid-Cyberinstitute

• NYS Grass-roots Effort
• Hauptman-Woodward Institute
• SUNY Binghamton
• SUNY Albany
• Columbia University
• WNY Grid NSF Funded
• SUNY Buffalo
• Hauptman-Woodward Institute
• Niagara University
• SUNY Geneseo
• Expand Virtual Facility

11
Center Infrastructure and Partners

• TeraGrid
• Science Gateway using ACDC Grid Portal
• UltraLight
• Configuring a small test cluster for Buffalo -
  UltraLight - CERN testing, 1GbE to 10 GbE
  connections
• NYSTAR
• Working with NYS support for research, education,
  and economic development, the New York State
  Office of Science, Technology and Academic
  Research
• GenNYsis Centers
• SUNY Albany and Geneseo University

12
Outline

• Primary Centers
• Enabling Technologies
• Modern Data Center Trends
• Enabling Services and Applications
• Enabling the next generation of discovery

13
ACDC Grid Portal Gateway
14
Grid-enabling Application Template (GAT)
15
Grid-enabling Application Template (GAT)
16
ACDC Data Grid Database Schema
ACDC Grid Data Grid
17
Data Grid API
18
ACDC Grid Dashboardhttp//osg.ccr.buffalo.edu
19
ACDC Operations Dashboardhttp//osg.ccr.buffalo.e
du
20
Grid Monitoring and Administration Collaborative
Environment (Grid MACE)
21
(No Transcript)
22
(No Transcript)
23
Outline

• Primary Centers
• Enabling Technologies
• Modern Data Center Trends
• Enabling Services and Applications
• Enabling the next generation of discovery

24
Modern Data Center

• Space
• 48U Server Racks
• Blade Centers, Multi-core Technology
• Virtual Machines
• Power
• AC vs DC Power
• 20-40 less thermal load and 30 less power
  consumption with DC power
• Cooling
• Data Center Thermal Management Infrastructure
• Floor mounted air handlers
• Server Rack Mounted Fans and Heat Exchangers
• Water Cooled Server Racks

25
Storage Trends Solutions

• Storage Area Network (SAN)
• Convenient, Failover Protection, Mid-level
  Performance
• Network Attached Storage (NAS)
• Dedicated, Failover Protection, High-level
  Performance
• Data Rates and File Usage
• Databases growing from 1 TB to 10s TB
• 3 year trending toward 64 growth rate per year
• 20 of filesystem files account for 80 of
  storage
• 3 year trending toward 75 growth rate per year
• 70 of files not touched 3 months after creation
• 40-50 of files not touched 6-12 months after
  creation

26
Storage Trends Solutions

• Move away from Direct Attached Storage (DAS)
• Use NAS or SAN solutions
• File System Archiving
• Email Archiving
• Eliminate multiple user email databases
• Single Instance Storage
• Eliminate multiple user document copies
• Smaller Backup Windows
• Less media
• Less infrastructure
• Reduce Primary Storage Using Tiered Approach
• Use filesystem guidelines for usage patterns
• Designate volatile storage areas
• Quota and scrub filesystems

27
Computational Trends Solutions

• xU Based Servers
• Required by science and engineering drivers
• x-core technology ready
• Blade Centers
• High density
• x-core technology ready
• Virtual Machines
• User can choose entire environment
• Operate multiple environments simultaneously
• Coprocessors and Field Programmable Gate Arrays
  (FPGAs)
• 3-10x more flops using less power

28
Outline

• Primary Centers
• Enabling Technologies
• Modern Data Center Trends
• Enabling Services and Applications
• Enabling the next generation of discovery

29
Domain Science and Engineering

• Shake-and-Bake (SnB)
• Molecular Structure Determination Application
• Buffalo-and-Pittsburgh (BnP)
• SnB and PHASES Complete Protein Phasing
• Q-Chem
• Commercial Quantum Chemistry Software Package
• SledgeHMMER
• Sequence analysis using profile hidden Markov
  models
• QMolDyn
• Quantum Molecular Dynamics
• ARACNE
• Algorithm for the Reconstruction of Accurate
  Cellular Networks

30
Domain Science and Engineering

• NWChem
• High Performance Computational Chemistry Software
• Ostrich
• Optimization and Parameter Estimation Tool for
  Groundwater Modeling
• Evolutionary Aseismic Design Retrofit (EADR)
• Passive Energy Dissipation System for Designing
  Earthquake Resilient Structures
• Princeton Ocean Model Great Lakes (POMGL)
• Great Lakes Hydrodynamic Circulation Model
• Titan
• Computational Modeling of Hazardous Geophysical
  Mass Flows

31
Outline

• Primary Centers
• Enabling Technologies
• Modern Data Center Trends
• Enabling Services and Applications
• Enabling the next generation of discovery

32
Enabling the Next Generation of Discovery

• Train the Next Generation Work Force
• Provide Ubiquitous Cyberinfrastructure
• Ensure Interoperability Through Standards Based
  Infrastructure
• Generate Virtual Facilities and Institutes
• Enable Collaborative Environments
• Obtain Efficient Utilization of Computational and
  Data Resources

33
Contact Information

• ACDC Grid and Operations Dashboards
• URL http//osg.ccr.buffalo.edu
• ACDC Grid Portal
• URL https//grid.ccr.buffalo.edu
• For GRASE VO Membership, email
• grid-help_at_ccr.buffalo.edu
• Center for Computational Research
• URL http//www.ccr.buffalo.edu
• URL http//www.ccr.buffalo.edu/grid
• General contact info
• Mark L. Green mlgreen_at_ccr.buffalo.edu