Download AJR - UB Computer Science and Engineering

Grid Computing Overview
Thanks to
Mark Ellisman
Data Acquisition
Advanced Visualization
Imaging Instruments
Analysis
Computational
Resources
Large-Scale Databases
 Coordinate Computing Resources, People, Instruments in Dynamic
Geographically-Distributed Multi-Institutional Environment
 Treat Computing Resources like Commodities
 Compute cycles, data storage, instruments
 Human communication environments
 No Central Control; No Trust
University at Buffalo The State University of New York
Center for Computational Research
CCR
Factors Enabling the Grid
 Internet is Infrastructure
Increased network bandwidth and advanced services
 Advances in Storage Capacity
Terabyte costs less than $5,000
 Internet-Aware Instruments
 Increased Availability of Compute Resources
Clusters, supercomputers, storage, visualization devices
 Advances in Application Concepts
Computational science: simulation and modeling
Collaborative environments  large and varied teams
 Grids Today
Moving towards production; Focus on middleware
University at Buffalo The State University of New York
Center for Computational Research
CCR
Computational Grids &
Electric Power Grids
 Similarities/Goals of CG and EPG
Ubiquitous
Consumer is comfortable with lack of
knowledge of details
 Differences Between CG and EPG
Wider spectrum of performance & services
Access governed by more complicated issues
Security
Performance
Socio-political factors
University at Buffalo The State University of New York
Center for Computational Research
CCR
Growth of Data and Load
Courtesy of
vs. Moore’s Law Rick Stevens
Metabolic Pathways
Pharmacogenomics
Human Genome
Combinatorial
Chemistry
Computational
Load
ESTs
Genome Data
Moore’s Law
1990
2000
University at Buffalo The State University of New York
2010
Center for Computational Research
CCR
A Short History of the Grid
 Grand Challenge Problems (1980s)
NSF and DOE initiatives
“Science is a team sport”
Initiate multi-resource projects involving computation,
instruments, visualization, data
 Evolution of Related Communities
Parallel computation
Address resource limitations
Networking
Gigabit testbed program
CASA Gigabit Testbed
(1990s)
 Investigate potential testbed network architectures
Explore usefulness for end-users
University at Buffalo The State University of New York
Center for Computational Research
CCR
The Globus Project
(Ian Foster and Carl Kesselman)
 Globus model focuses The Grid as a Layered Set of Services
on providing key
Applications
Grid services
 Resource access and
management
 Grid FTP
 Information Service
 Security services
Authentication
Authorization
Policy
Delegation
 Network reservation,
monitoring, control
High-level Services and Tools
GlobusView
DUROC
MPI
MPI-IO
CC++
Testbed Status
Nimrod/G
globusrun
Core Services
Nexus
Metacomputing
Directory Service
Gloperf
Condor
LSF
Heartbeat
Monitor
Local
Services
MPI
Easy
GRAM
Globus
Security
Interface
NQE
University at Buffalo The State University of New York
AIX
GASS
TCP
UDP
Irix
Solaris
Center for Computational Research
CCR
NSF Extensible TeraGrid Facility
Caltech: Data collection analysis
0.4 TF IA-64
IA32 Datawulf
80 TB Storage
Sun
IA64
ANL: Visualization
LEGEND
Cluster
Visualization
Cluster
Storage Server
Shared Memory
IA32
IA64
IA32
Disk Storage
Backplane Router
1.25 TF IA-64
96 Viz nodes
20 TB Storage
IA32
Extensible Backplane Network
LA
Hub
30 Gb/s
Chicago
Hub
40 Gb/s
30 Gb/s
30 Gb/s
4 TF IA-64
DB2, Oracle Servers
500 TB Disk Storage
6 PB Tape Storage
1.1 TF Power4
IA64
10 TF IA-64
128 large memory nodes
230 TB Disk Storage
GPFS and data mining
Sun
30 Gb/s
30 Gb/s
Figure courtesy of
Rob Pennington, NCSA
EV7
6 TF EV68
71 TB Storage
0.3 TF EV7 shared-memory
150 TB Storage Server
IA64
EV68
Pwr4
SDSC: Data Intensive
NCSA: Compute Intensive
University at Buffalo The State University of New York
Sun
PSC: Compute Intensive
Center for Computational Research
CCR
Critical Resources:
WNY Computational & Data Grids
 Computational & Data Resources (CCR)
10TF Computing & 78TB Storage
 Instruments (HWI, RPCI)
Microarray; Diffractometer; NMR
High-Throughput Crystallization Laboratory
 Data Generation (HWI)
7TB per year
 Databases (UB-N, UB-S, BGH, CoE)
SnB; Multiple Sclerosis; Protein/Genomic
University at Buffalo The State University of New York
Center for Computational Research
CCR
Network Connections
1000 Mbps
100 Mbps
FDDI
100 Mbps
1.54 Mbps (T1) - RPCI
OC-3 - I1
Medical/Dental
44.7 Mbps (T3) - BCOEB
155 Mbps (OC-3) I2
1.54 Mbps (T1) - HWI
NYSERNet
NYSERNet
350
Main St
350 Main St
BCOEB
Abilene
622 Mbps (OC-12)
University at Buffalo The State University of New York
Commercial
Center for Computational Research
CCR
Network Connections (New)
1000 Mbps
100 Mbps
100 Mbps
FDDI
Medical/Dental
RIA
RIA
UB controlled
UB controlled
meet-me
loc.
meet-me loc.
OC-3 - I1
1000 Mbps
1000 Mbps
BCOEB
1000 Mbps
Abilene
University at Buffalo The State University of New York
NYSERNet
NYSERNet
350
Main St
350 Main St
622 Mbps (OC-12)
Commercial
Center for Computational Research
CCR
Advanced CCR Data Center (ACDC)
Computational Grid Overview
Nash: Compute Cluster
75 Dual Processor
1 GHz Pentium III
RedHat Linux 7.3
1.8 TB Scratch Space
Joplin: Compute Cluster
300 Dual Processor
2.4 GHz Intel Xeon
RedHat Linux 7.3
38.7 TB Scratch Space
Mama: Compute Cluster
9 Dual Processor
1 GHz Pentium III
RedHat Linux 7.3
315 GB Scratch Space
ACDC: Grid Portal
4 Processor Dell 6650
1.6 GHz Intel Xeon
RedHat Linux 9.0
66 GB Scratch Space
Young: Compute Cluster
16 Dual Sun Blades
47 Sun Ultra5
Solaris 8
770 GB Scratch Space
SGI Origin 3800
64 - 400 MHz IP35
IRIX 6.5.14m
360 GB Scratch Space
Fogerty: Condor Flock Master
1 Dual Processor
250 MHz IP30
IRIX 6.5
Expanding
RedHat, IRIX, Solaris,
WINNT, etc
CCR
T1 Connection
Computer Science & Engineering
25 Single Processor Sun Ultra5s
Crosby: Compute Cluster
19 IRIX, RedHat, &
WINNT Processors
School of Dental Medicine
Hauptman-Woodward Institute
9 Single Processor Dell P4 Desktops
13 Various SGI IRIX Processors
Note: Network connections are 100 Mbps unless otherwise noted.
University at Buffalo The State University of New York
Center for Computational Research
CCR
ACDC Data Grid Overview
182 GB Storage
Joplin: Compute Cluster
300 Dual Processor
2.4 GHz Intel Xeon
RedHat Linux 7.3
38.7 TB Scratch Space
Nash: Compute Cluster
75 Dual Processor
1 GHz Pentium III
RedHat Linux 7.3
1.8 TB Scratch Space
100 GB Storage
70 GB Storage
Mama: Compute Cluster
9 Dual Processor
1 GHz Pentium III
RedHat Linux 7.3
315 GB Scratch Space
ACDC: Grid Portal
56 GB Storage
Young: Compute Cluster
4 Processor Dell 6650
1.6 GHz Intel Xeon
RedHat Linux 9.0
66 GB Scratch Space
16 Dual Sun Blades
47 Sun Ultra5
Solaris 8
770 GB Scratch Space
CSE Multi-Store
2 TB
100 GB Storage
Crosby: Compute Cluster
136 GB Storage
SGI Origin 3800
64 - 400 MHz IP35
IRIX 6.5.14m
360 GB Scratch Space
Network Attached
Storage
480 GB
Storage Area Network
75 TB
Note: Network connections are 100 Mbps unless otherwise noted.
University at Buffalo The State University of New York
Center for Computational Research
CCR
WNY Grid Highlights
 Heterogeneous Computational & Data Grid
 Currently in Beta with Shake-and-Bake
 WNY Release in March
 Bottom-Up General Purpose Implemenation
Ease-of-Use User Tools
Administrative Tools
 Back-End Intelligence
Backfill Operations
Prediction and Analysis of Resources to Run
Jobs (Compute Nodes + Requisite Data)
University at Buffalo The State University of New York
Center for Computational Research
CCR
Advanced CCR Data Center (ACDC)
Computational Grid Overview
Nash: Compute Cluster
75 Dual Processor
1 GHz Pentium III
RedHat Linux 7.3
1.8 TB Scratch Space
Joplin: Compute Cluster
300 Dual Processor
2.4 GHz Intel Xeon
RedHat Linux 7.3
38.7 TB Scratch Space
Mama: Compute Cluster
9 Dual Processor
1 GHz Pentium III
RedHat Linux 7.3
315 GB Scratch Space
ACDC: Grid Portal
4 Processor Dell 6650
1.6 GHz Intel Xeon
RedHat Linux 9.0
66 GB Scratch Space
Young: Compute Cluster
16 Dual Sun Blades
47 Sun Ultra5
Solaris 8
770 GB Scratch Space
SGI Origin 3800
64 - 400 MHz IP35
IRIX 6.5.14m
360 GB Scratch Space
Fogerty: Condor Flock Master
1 Dual Processor
250 MHz IP30
IRIX 6.5
Expanding
RedHat, IRIX, Solaris,
WINNT, etc
CCR
T1 Connection
Computer Science & Engineering
25 Single Processor Sun Ultra5s
Crosby: Compute Cluster
19 IRIX, RedHat, &
WINNT Processors
School of Dental Medicine
Hauptman-Woodward Institute
9 Single Processor Dell P4 Desktops
13 Various SGI IRIX Processors
Note: Network connections are 100 Mbps unless otherwise noted.
University at Buffalo The State University of New York
Center for Computational Research
CCR
Data Grid Motivation & Goal
 Motivation:
 Large data collections are emerging as important
community resources.
 Data Grids inherently complements Computational
Grids, which manipulate data.
 A data grid denotes a large network of distributed storage
resources such as archival systems, caches, and databases,
which are linked logically to create a sense of global
persistence.
 Goal:
 To design and implement transparent management of
data distributed across heterogeneous resources, such
that the data is accessible via a uniform web interface.
University at Buffalo The State University of New York
Center for Computational Research
CCR
Data Grid Summary
 544 GB Storage
 Located on 6 heterogeneous
ACDC-Grid resources
 480 GB Storage
 Located on 1 dual processor
Dell PowerVault server
 75,000 GB Storage (10/03)
 Served by 4 – 16 processor HP
GS1280 servers
 2,000 GB Storage
Network Attached
Storage
480 GB
56 GB Storage
70 GB Storage
100 GB Storage
100 GB Storage
CSE Multi-Store
2 TB
136 GB Storage
182 GB Storage
 Served by Sun Ultra-60 servers
 78,024 GB Total Data Grid
Storage available and accessible
from the ACDC-Grid Portal
University at Buffalo The State University of New York
Storage Area Network
75 TB
Center for Computational Research
CCR
Grid-Based SnB
Objectives
 Install Grid-Enabled Version of SnB
 Job Submission and Monitoring over Internet
 SnB Output Stored in Database
 SnB Output Mined through Internet-Based
Integrated Querying Tool
 Serve as Template for Chem-Grid & Bio-Grid
 Experience with Globus and Related Tools
University at Buffalo The State University of New York
Center for Computational Research
CCR
Grid Enabled SnB
 Problem Statement
 Use all available resources in the ACDC-Grid for determining a
single molecular structure.
 Grid Enabling Criteria
 All heterogeneous resources in the ACDC-Grid are capable of
executing the SnB application.
 All job results obtained from the ACDC-Grid resources are
stored in a corresponding molecular structure database.
 There are three modes of operation:
 Continue submitting SnB application jobs until
 the grid-enabled SnB application determines a solution has been found, or
 “X” number of trials have been evaluated, or
 indefinitely (grid job owner determines when a solution has been found).
University at Buffalo The State University of New York
Center for Computational Research
CCR
Grid Services and Applications
Applications
Shake-and-Bake
ACDC-Grid
Computational
Resources
High-level Services
and Tools
Globus
Toolkit
MPI
Apache
MPI-IO
C, C++, Fortran, PHP
Oracle
MySQL
NWS
globusrun
Core Services
Metacomputing
Directory
Service
Condor
LSF
Globus
Security
Interface
Local Services
Stork
MPI
PBS
Maui Scheduler
TCP
UDP
GRAM
GASS
RedHat Linux
WINNT
Irix
ACDC-Grid
Data
Resources
Solaris
Adapted from Ian Foster and Carl Kesselman
University at Buffalo The State University of New York
Center for Computational Research
CCR
Notes
 Apache – web portal server
 PHP - used by apache server for dynamic web
portal pages
 MDS – traditional to use MDS with LDAP but we
use MDS with MYSql grid portal database to keep
information of available resources (we poll every
15 mins)
 GRAM – Globus Resource Allocation Manager –
API for requesting comptuational jobs
 GASS – Global Access to Secondary Storage – API
for accessing files stored on various platforms
 Stork – Condor module for transporting job files
within a flock
CCR
University at Buffalo The State University of New York
Center for Computational Research
Grid Enabled SnB
 Required Layered Grid Services
 Grid-enabled Application Layer
 Shake – and – Bake application
 Apache web server
 MySQL database
 High-level Service Layer
 Globus, NWS, PHP, Fortran, and C
 Core Service Layer
 Metacomputing Directory Service, Globus Security Interface,
GRAM, GASS
 Local Service Layer
 Condor, MPI, PBS, Maui, WINNT, IRIX, Solaris, RedHat Linux
University at Buffalo The State University of New York
Center for Computational Research
CCR
Required Grid Services
 Application Layer
 Shake-and-Bake
 Apache web server
 MySQL database
 High-level Services
Grid Implementation as a Layered Set of Services
Applications
High-level Services and Tools
GlobusView
DUROC
MPI
MPI-IO
CC++
Testbed Status
Nimrod/G
globusrun
 Globus, PHP, Fortran, C
 Core Services
 Metacomputing Directory
Service, Globus Security
Interface, GRAM, GASS
Core Services
Nexus
Metacomputing
Directory
Service
Gloperf
 Local Services
 Condor, MPI, PBS, Maui,
WINNT, IRIX, Solaris,
RedHat Linux
Condor
LSF
University at Buffalo The State University of New York
GRAM
Heartbeat
Monitor
Local
Services
MPI
Easy
Globus
Security
Interface
NQE
AIX
GASS
TCP
UDP
Irix
Solaris
Center for Computational Research
CCR
Grid Enabled SnB Execution
User
defines Grid-enabled SnB job using Grid Portal or SnB
supplies location of data files from Data Grid
supplies SnB mode of operation
Grid Portal
assembles required SnB data and supporting files,
execution scripts, database tables.
determines available ACDC-Grid resources.
ACDC-Grid job management includes:
automatic determination of appropriate execution times,
number of trials, and number/location of processors,
logging/status of concurrently executing resource jobs, &
automatic incorporation of SnB trial results into the
molecular structure database.
University at Buffalo The State University of New York
Center for Computational Research CCR
ACDC-Grid Portal
University at Buffalo The State University of New York
Center for Computational Research
CCR
ACDC-Grid Portal Login
Grid Portal
login
screen
University at Buffalo The State University of New York
Center for Computational Research
CCR
Data Grid Capabilities
Browser view of
“mlgreen” user
files stored in the
Data Grid
University at Buffalo The State University of New York
Center for Computational Research
CCR
Data Grid Capabilities
Browser view of
“miller” group files
published by user
“rappleye”
University at Buffalo The State University of New York
Center for Computational Research
CCR
Data Grid Capabilities
Browser view of
“public” user
files published
by user “miller”
University at Buffalo The State University of New York
Center for Computational Research
CCR
Data Grid Capabilities
University at Buffalo The State University of New York
Center for Computational Research
CCR
Data Grid Capabilities
University at Buffalo The State University of New York
Center for Computational Research
CCR
Grid Portal Job Status
 Grid-enabled jobs can be
monitored using the Grid Portal
web interface dynamically.
 Charts are based on:
total CPU hours, or
total jobs, or
total runtime.
 Usage data for:
running jobs, or
queued jobs.
 Individual or all resources.
 Grouped by:
group, or
user, or
queue.
University at Buffalo The State University of New York
Center for Computational Research
CCR
Grid Portal Job Status
University at Buffalo The State University of New York
Center for Computational Research
CCR
ACDC-Grid Portal Condor Flock
 CondorView
integrated into
ACDC-Grid Portal
University at Buffalo The State University of New York
Center for Computational Research
CCR
ACDC-Grid Portal User
Management
Administrator
based
University at Buffalo The State University of New York
user based
Center for Computational Research
CCR
ACDC-Grid Portal
Resource Management
 Administrator grants a user
access to ACDC-Grid
 resources,
 software, and
 web pages.
University at Buffalo The State University of New York
Center for Computational Research
CCR
ACDC-Grid Administration
University at Buffalo The State University of New York
Center for Computational Research
CCR
ACDC-Grid Administration
University at Buffalo The State University of New York
Center for Computational Research
CCR
Grid Enabled Data Mining
 Problem Statement
Use all available resources in the ACDC-Grid for
executing a data mining genetic algorithm
optimization of SnB parameters for molecular
structures having the same space group.
 Grid Enabling Criteria
All heterogeneous resources in the ACDC-Grid are
capable of executing the SnB application.
All job results obtained from the ACDC-Grid
resources are stored in a corresponding molecular
structure databases.
University at Buffalo The State University of New York
Center for Computational Research
CCR
Grid Enabled Data Mining
 There are two modes of operation and two sets of
stopping criteria:
 Data mining jobs can be submitted in
 a dedicated mode (time critical), where jobs are queued on
ACDC-Grid resources, or
 in a back fill mode (non-time critical), where jobs are submitted
to ACDC-Grid resource that have unused cycles available.
 There are two sets of stopping criteria:
 Continue submitting SnB data mining application jobs
until
 the grid-enabled SnB application determines optimal parameters
have been found, or
 indefinitely (grid job owner determines when optimal
parameters have been found).
University at Buffalo The State University of New York
Center for Computational Research
CCR
Grid Enabled Data Mining
ACDC-Grid
Data Grid
Data
Mining
Criteria
ACDC-Grid Computational
Resources
Grid Portal
Workflow Job
Manager
University at Buffalo The State University of New York
Molecular
Structure
Database
Center for Computational Research
CCR
SnB Molecular Structure
Database
Molecular
Structure
Database
University at Buffalo The State University of New York
Center for Computational Research
CCR
Grid Enabled Data Mining
 Execution Scenario
 User defines a Grid-enabled data mining SnB job using the Grid
Portal web interface supplying:
 designate which molecular structures parameter sets to optimize,
 data file metadata, and
 Grid-enabled SnB mode of operation dedicated or back fill mode, and
 Grid-enabled SnB stopping criteria.
 The Grid Portal assembles the required SnB application data and
supporting files, execution scripts, database tables, and submits jobs
for parameter optimization based on the current database statistics.
 ACDC-Grid job management includes:
 automatic determination of appropriate execution times, number of trials,
and number of processors for each available resource,
 logging and status of all concurrently executing resource jobs,
 automatic incorporation of SnB trial results into the molecular structure
database, and
 post processing of updated database for subsequent job submissions.
University at Buffalo The State University of New York
Center for Computational Research
CCR
ACDC Data Grid Database
Schema
ACDC-Grid
Data Grid
University at Buffalo The State University of New York
Center for Computational Research
CCR
Grid Portal Job Status
ACDC-Grid
Computational
Resources
University at Buffalo The State University of New York
Center for Computational Research
CCR
Data Grid Overview
 Enable the transparent migration of data
between various resources while preserving
uniform access for the user.
Maintain metadata information about each file
and its location in a global database table.
Currently using MySQL tables.
Periodically migrate files between machines for
more optimal usage of resources.
University at Buffalo The State University of New York
Center for Computational Research
CCR
Data Grid Functionality
 Implement basic file management functions
accessible via a platform-independent web
interface.
 Features include:
 User-friendly menus/ interface.
 File Upload/ Download to and from the Data Grid
Portal.
 Simple web-based file editor.
 Efficient search utility.
 Logical display of files for a given user in three divisions
(user/ group/ public).
Hierarchical vs. List-based
3 divisions: (user/ group/ public)
Sorting capability based on file metadata, i.e. filename,
size, modification time, etc.
University at Buffalo The State University of New York
Center for Computational Research
CCR
Data Grid Functionality
 Support multiple access to files in the data
grid.
Implement basic Locking and Synchronization
primitives for version control.
 Integrate security into the data grid.
Implement basic authentication and
authorization of users.
Decide and enforce policies for data access and
publishing.
University at Buffalo The State University of New York
Center for Computational Research
CCR
Data Grid File Migration
 Migration Algorithm
File migration depends upon a number of
factors:
User access time
Network capacity at time of migration
User profile
User disk quotas on various resources
University at Buffalo The State University of New York
Center for Computational Research
CCR
Data Grid File Migration
 We need to mine log files in order to
determine
How much data to migrate in one migration
cycle?
What is an appropriate migration cycle length?
What is a user’s access pattern for files?
What is the overall access pattern for particular
files?
University at Buffalo The State University of New York
Center for Computational Research
CCR
Data Grid File Aging
 Global File Aging vs. Local File Aging
User aging attribute
Indicative of a user’s access across their own
files.
Attribute of a user’s profile.
During migration time, this attribute will
determine which user’s files should be migrated
off of the grid portal onto a remote resource.
Function of (file age, global file aging, resource
usage)
University at Buffalo The State University of New York
Center for Computational Research
CCR
Data Grid File Aging
 File aging attribute
 Indicative of overall access to/migration activity of a
particular file.
 Attribute in file_management table.
 Scale: 0 to 1 probability of whether or not to migrate file.
 File_aging_local_param initialized to 1.
 During migration time after a user has been chosen, this
attribute will help determine which files of the user to
migrate.
i.e. Migrate a maximum of the top 5% of user’s files in any
one cycle.
University at Buffalo The State University of New York
Center for Computational Research
CCR
Data Grid File Aging
 For a given user, the average of the
file_aging_local_param attributes of all files should
be close to 1.
 Operating tolerance before action is taken is within the
range of 0.9 – 1.1.
 In this way, the user file_aging_global_param can
be a function of this average.
 If the average file_aging_local_param attribute > 1.1,
then files of the user are being held to long before being
migrated.
The file_aging_global_param value should be decreased.
 If the average file_aging_local_param attribute < 0.9,
then files of the user are being accessed at a higher
frequency than the file_aging_global_param value.
The file_aging_global_param value should be increased.
University at Buffalo The State University of New York
Center for Computational Research
CCR
Data Grid Resource Info
University at Buffalo The State University of New York
Center for Computational Research
CCR
Data Grid Resource Info
Both platforms have
reduced bandwidth
available for additional
transfers
University at Buffalo The State University of New York
Center for Computational Research
CCR
Date Grid File Management
Table
University at Buffalo The State University of New York
Center for Computational Research
CCR
Data Grid File Age
 File age, access time,
and resource id
denote:
 the amount of time
since a file was
accessed,
 when the file was
accessed, and
 where the file
currently resides
respectively.
University at Buffalo The State University of New York
Center for Computational Research
CCR
Data Grid Summary
Network Attached
Storage
480 GB
56 GB Storage
70 GB Storage
100 GB Storage
100 GB Storage
CSE Multi-store
2 TB
136 GB Storage
182 GB Storage
Storage Area Network
75 TB
 The Data Grid
algorithms are
continually evolving to
minimize network
traffic and maximize
disk space utilization
on a per user basis by
data mining user
usage and disk space
requirements.
University at Buffalo The State University of New York
Center for Computational Research
CCR
ACDC-Grid
Development/Maintenance
 Development Requirements
 7 – Person months for Grid
Services Coordinator
 Minimum Maintenance
Requirements
 Including Grid and Database
conceptual design and
implementation
 5 – Person months for Grid
Services Programmer
 Web portal programming
 5 – Person months for System
Administrator
 Globus, NWS, MDS, etc.
installations
 3 – Person months for Database
Administrator
 Grid Portal Database
implementation
University at Buffalo The State University of New York
 1 – Grid Services
Coordinator
100% level of effort
 1 – Grid Services
Programmer
100% level of effort
 1 – System Administrator
50% level of effort
 1 – Database Administrator
10% level of effort
Center for Computational Research
CCR
Future ACDC Applications
 Princeton Ocean Model (POM)
 Genetic Algorithms for Earthquake
Structural Design
 Bioinformatics
 Computational Chemistry (Q-Chem)
 Environmental Engineering Applications
University at Buffalo The State University of New York
Center for Computational Research
CCR