IWB Briefing Materials

1
IBM Grid Vision Strategy
GFK 2002 Summer Workshop
IBM Korea
2
The Internet - mid '90's
. . . .
Information
World Wide Web
Communications
e-mail
Networking
TCP/IP

3
Current IT Environment
Distributed, Heterogeneous, Complex
Middleware
Directory and Security Servers
Transaction Servers
Customers
Business Partners
Network
Quality of Service
Web Presentation Servers
Web Application Servers
Data Servers
Suppliers
Employees
Storage
4
Internet Evolution
Distributed
Computing
The Grid
Information
World Wide Web
Communications
e-mail
Networking
TCP/IP

5
Grid Computing
Distributed Computing Over the Internet Using
Open Standards

• Virtual, collaborative organizations sharing
  applications and data in an open heterogeneous
  environment
• A vast aggregation of geographically dispersed
  computing resources

Virtual Servers, Storage and Instruments
Grid Middleware
Distributed Physical Servers and Storage
6
Grid Computing
Microcosm Pre-Internet System
Data
Storage
Applications
Processing
I/O
Operating System
7
Grid Computing
Macrocosm Distributed Resources and
Applications
Data
Storage
Applications
Processing
I/O
Operating System
....a single unified image
8
Grids Enable

• Sharing of Resources -

Virtual Organizations Collaboration

• Management of Resources -

Quality of Service Optimization

• Access to Resources -

On Demand Computing and Utility Models
9
Open Grid Services Architecture
Applications
Middleware
Systems Management and Automation
Workload / Performance Management
Security
Availability / Service Management
Logical Resource Management
Clustering Services
Connectivity Management
Physical Resource Management

OS
OS
OS
10
Architecture Framework
http//www.globus.org/research/papers/ogsa.pdf
STC
xSP
etc...
eCommerce
End-Users Admin Staff
End-Users
New classes of applications enabled

Common GUI
Autonomic Functions Management Middleware
Open Grid Services Architecture
WebServices Run-time
IBM Platforms QoS Enablers/Optimizers
Solaris HP/UX
Linux
Windows
...
Storage
z/OS
Windows
Linux
AIX
OS/400
zSeries
pSeries
iSeries
z,p,i,xSeries
xSeries
Sun HP
DellCompaq ...
DellCompaq ...
eLiza eServer Foundation
11
The Best of Both Worlds
Web Services Grid Protocols
Open Grid Services Architecture
manage
share
access
Continuous Availability
Applications on demand
Resources on demand
Secure and universal access
Global accessibility
Business integration
Vast resource scalability
Web Services
Grid Protocols
12
Uses of Grid Technology
4 Models, Unique Value Propositions
Processing
Aggregate processing power from a distributed
collection of heterogeneous systems
Customer Values
Partners
Increased
Decreased
Platform, Entropia, United Devices

• Productivity
• Flexibility
• Resource use
• Reliability/Availability

• Complexity
• Total cost of ownership

13
Uses of Grid Technology
4 Models, Unique Value Propositions
Data
Secure access and sharing of distributed data
information ina collaborative fashion
Processing
Aggregate processing power from a distributed
collection of heterogeneous systems
Partners
Avaki, Platform
Customer Values
Partners
Increased
Decreased
Platform, Entropia, United Devices

• Productivity
• Flexibility
• Resource use
• Reliability/Availability

• Complexity
• Total cost of ownership

14
Uses of Grid Technology
4 Models, Unique Value Propositions
Resiliency
Improve the quality of service of distributed
systems, despite unplanned events
Data
Secure access and sharing of distributed data
information ina collaborative fashion
Processing
Partners
Aggregate processing power from a distributed
collection of heterogeneous systems
Platform
Partners
Avaki, Platform
Customer Values
Partners
Increased
Decreased
Platform, Entropia, United Devices

• Productivity
• Flexibility
• Resource use
• Reliability/Availability

• Complexity
• Total cost of ownership

15
Uses of Grid Technology
4 Models, Unique Value Propositions
On Demand
Access data processing capabilities in a
utility-like fashion.. Make vs. Buy
Resiliency
Improve the quality of service of distributed
systems, despite unplanned events
Data
Partners
Platform, Entropia, United Devices, Avaki
Secure access and sharing of distributed data
information ina collaborative fashion
Processing
Partners
Aggregate processing power from a distributed
collection of heterogeneous systems
Platform
Partners
Avaki, Platform
Client Values
Partners
Increased
Decreased
Platform, Entropia, United Devices

• Resource use
• Flexibility
• Productivity
• Reliability/Availability

• Complexity
• Total cost of ownership

16
Grid Deployment Options
A function of business need, technology and
organizational flexibility
1. Intra-Grids
Grid
Grid
NAS/SAN
NAS/SAN
17
Grid Deployment Options
A function of business need, technology and
organizational flexibility
2. Extra-Grids
1. Intra-Grids
Grid
Grid
Grid
VPN
NAS/SAN
NAS/SAN
NAS/SAN
18
Grid Deployment Options
A function of business need, technology and
organizational flexibility
3. Inter-Grids
2. Extra-Grids
1. Intra-Grids
Cactus
(SF) Express Project
NTG
Grid
Grid
Grid
VPN
NAS/SAN
NAS/SAN
NAS/SAN
Fin.Services
MFG
19
Grid Adoption Framework
Several Logical Paths to Implementation
1
2
3
Grid Application Services Server (GASS)

• xSP
• Utility Trading/ Brokering Hubs

On Demand
D
Business and IT Play
Grid Services Server (GSS)
Resiliency
C
Increasing IT Complexity
Grid Edge Server - Data (GES)
Data
B
IT Play
Processing
A
Grid Management Point (GMP)
Grid Switching Point (GSP)
Grid Point of Presence (GPOP)
Grid Trading Point (GTP)
Intra Grid
Extra Grid
Inter Grid
Increasing Organization Complexity
20
Early Adopters
e-Science Grids

• UK National Grid
• Netherlands National Grid
• The TeraGrid
• University of Pennsylvania
• North Carolina BioGrid
• Department of Energy Science Grid

21
The TeraGrid

• Heterogeneous Systems
• National Center for Supercomputing Applications
• San Diego Supercomputing Center
• Argonne National Laboratory
• California Institute of Technology

13.6 trillion floating point operations per
second 600 terabytes of data 40 gigabits per
second Accessible to thousands of scientists
working on advanced research
22
UK Research Grid

• Collaborative, multidisciplinary, scientific
  research
• Test bed for utility computing
• Future commercial applications

EU
Edinburgh
US Sites
CERN
Glasgow
Newcastle
Belfast
Dublin
Manchester
Cambridge
Oxford
Cardiff
London
Southampton
23
National Digital Mammographic Archive
An Archive and Data Utility for Medical and
Diagnostic Content
U Toronto
CANet2
CANet2
New York GigaPop
U CHI
MAGPI
MREN
STARTAP
Chicago Peering Point
U Penn
NGIX
Indianapolis GigaPop
Abilene
Peer Network
Oak Ridge
U NC
Abilene Connector
ESNet
Project Site
NCNI
SOX
Atlanta GigaPop
24
Indiana Virtual Machine

• Purdue and Indiana University Supercomputers
• 1.4 teraflops
• High-speed optical fiber network
• Sharing cycles and capacity
• Ultra-large calculations
• Simulating consumerbehavior
• Simulating homeland security

25
Butterfly.net

• Unlimited Numbers of Players
• Distributed Artificial Intelligence
• Multiple Concurrent Players
• 1,000 downloads of developers kit per week
• Hot-swappable Components
• Developers, Publishers, ESPs

26
Commercial Grid Computing Drivers

• Maximize utilization of existing resources
• Simplify operating environment
• Improve availability productivity
• Enable collaboration virtual organizations
• Enhance promote flexibility

....and enabled by Self-managingTechnologies
27
Commercializing the Grid - Considerations
Enterprise Quality of Service Across the Grid
Security
Heterogeneous Workload Management
Governance Policy Standards
Disaster Recovery
Autonomic Functionality
End-to-End Systems Management
Accounting Service Level Agreements
28

On Demand Computing
Realized Value
Quality of Service
Variable Cost Flexibility
Make vs.Buy Option
Reduced Complexity
Greater Productivity
29
Tapping The Power of Grid Computing