GRID Computing: On the path

to utility computing
-Dr. Srinivas Padmanabhuni
SETLabs

28th Nov 2003 GRID COMPUTING 1

 Agenda
• Why GRID?
• What is GRID?
• Where GRID applies?
• How is GRID constructed?
• OGSA = Standards based GRID
• Conclusions and Call For Action

28th Nov 2003 GRID COMPUTING 2

 Why GRID?

28th Nov 2003 GRID COMPUTING 3

 Today’s IT architecture is complex and
unmanageable…
internet internet
internet internet

access tier edge routers

routing
switches
authentication, DNS,
intrusion detect, VPN 1st level firewall
web cache

web tier load balancing

switches
web
servers
web page storage
(NAS) 2nd level firewall

application tier switches

application
servers
files
(NAS) switches

database tier database

SQL servers

storage area
network
(SAN)

28th Nov 2003 GRID COMPUTING 4

 Moore’s Law for Computing Speed..
10 GHz by 2010
10,000
Pentium® 4 Processor

1,000
Pentium® II Processor
Pentium® III Processor

100 486™ Processor

Pentium® Processor
MHz
286 386™ Processor
10 8085

4004 8086
1 8080

0.1

1970 1980 1990 2000 2010

28th Nov 2003 Source: Intel

GRID COMPUTING 5
 Network is growing even faster…
• Network vs. computer performance
–Computer speed doubles every 18 months
–Network speed doubles every 9 months
–Difference = order of magnitude per 5 years
• 1986 to 2000
–Computers: x 500
–Networks: x 340,000
• 2001 to 2010
–Computers: x 60
–Networks: x 4000
Moore’s Law vs. storage improvements vs. optical improvements. Graph from Scientific American
(Jan-2001) by Cleo Vilett, source Vined Khoslan, Kleiner, Caufield and Perkins.
28th Nov 2003 GRID COMPUTING 6
Putting together network and computing
speed rates of growth..
• Moore’s Law. 1,000,000,000,000
– Transistors on a single chip 100,000,000,000
doubles approximately every 10,000,000,000
18–24 months.
1,000,000,000
Moore/
100,000,000 Transistors
• Gilder’s Law. 10,000,000
– Aggregate bandwidth triples 1,000,000
approximately every year.
100,000
10,000 Metcalf/
• Metcalfe’s Law. 1,000 Gilder/
Network
Nodes
– The value of a network may Bandwidth
grow exponentially with the 100
number of participants. 10
1
1970 1975 1980 1985 1990 1995 2000 2005 2010
2,300 6,000 29,000 275,000 1.2 mil 5.5 mil 42 mil 252 mil 1.344 bil

50 50 56 1,544 45,000 145,000 10 mil 2.43 bil 200.49 bil

4 111 200 10,000 300,000 1 mil 140 mil 3.5 bil 300 bil

Source: Cambridge Energy Resource Associates

10616-177
28th Nov 2003 GRID COMPUTING
 Low Infrastructure Utilization in today’s IT
architecture..
Peak-hour Prime-shift 24-hour Period
Utilization Utilization Utilization
Mainfram
85-100% 70% 60%
es
UNIX 50-70% 10-15% <10%
Intel-
30% 5-10% 2-5%
based
Storage N/A N/A 52%
Source: IBM Scorpion White Paper: Simplifying the Corporate IT Infrastructure, 2000

28th Nov 2003 GRID COMPUTING 8

 To summarize…
• Moore’s law improvements in computing produce highly
functional end-systems
• The Internet and burgeoning wired and wireless provide
universal connectivity
• Collaborative modes of working and problem solving
emphasize teamwork, computation
• Network exponentials produce dramatic changes in
geometry and geography
• Standards for application to application communication
getting universal acceptance
• Pressure on effective utilization of resources in
enterprises due to current low utilization rates..

28th Nov 2003 GRID COMPUTING 9

 Enter on-demand computing and
related concepts…
• On-Demand Computing: A conglomerate of multiple concepts
to enable respond to elastic computing demand using inelastic computing
resources
– IBM calls it e-business On-Demand, Sun calls it N1, HP calls it adaptive
management, Microsoft calls it the Dynamic Systems Initiative
• Utility Computing: Computing Resources made available (like
electricity etc.) as needed, and charged on usage
• Autonomic Computing: A self-managing computing model
where computing resources are controlled without human intervention,
includes self-healing, self-protecting, self-optimizing, and self-configuring
features.
• GRID Computing: Pooling of multiple resources coordinated to
appear as a single virtual resource to the external world

28th Nov 2003 GRID COMPUTING 10

 What is GRID?

28th Nov 2003 GRID COMPUTING 11

 A story first...

The pigeons showed that..

Unity is Strength
28th Nov 2003 GRID COMPUTING 12
 Defining GRID..
"A computational grid is a hardware and software
infrastructure that provides dependable,
consistent, pervasive, and inexpensive access
to high-end computational capabilities.”
-The Grid: Blueprint for a New Computing Infrastructure;
Morgan Kaufmann; San Francisco; 1999

“Flexible, secure, coordinated resource sharing

among dynamic collections (VOs) of individuals,
institutions, and resources”
-The Anatomy of the Grid: Enabling Scalable Virtual Organizations

28th Nov 2003 GRID COMPUTING 13

 What constitutes GRID?
– Coordination of resources that are not subject to
centralized control …
– Heterogeneous mix of resources…
– Loosely Coupled connections ..
– Flexible and Dynamic Collection..
– Usage of standard, open, general-purpose protocols
and interfaces …
– Virtualization of resources to create VOs..
– Non-Trivial quality of service…
– Resources may be processors, computers, clusters,
data, databases, scientific instruments, displays, etc..

28th Nov 2003 GRID COMPUTING 14

 What is not a GRID?
– A cluster of homogeneous machines..
– A network attached storage device..
– A standalone scientific instrument..
– A huge standalone supercomputer..
– A Massively Parallel Processing computer..
– A multi-processor computer..
– A high-speed network..
– A homogeneous cluster of computers
Each is an important component of a Grid,
but by itself does not constitute a Grid

28th Nov 2003 GRID COMPUTING 15

 A Picture of GRID…

28th Nov 2003 GRID COMPUTING 16

 History of the GRID
• 1986 - First Connection Machine CM-1 • Jan 2001 - EU DataGrid project launch
• 1987 - First CM-2 • Mar 2001 - Global Grid Forum 1
• 1988 - Condor project begins
• 1990 - PVM project begins • Jul 2001 - UK e-Science Programme launch
• 1991 - WWW created by Tim Berners-Lee at CERN • Aug 2001 - US TeraGrid project launch
• 1991 - UK JANet goes IP
• 1991 - nCUBE running Oracle PS achieves 1,037 Tps (2x mainframe
• Nov 2001 - GEANT, the pan-EU gigabit network,
speed, 0.05x cost) activated
• 1992 - CODINE project underway • 2002
• 1993 - First Cray T3D
• 1993 – Legion, a GRID Object model project launch – Dozens of application communities &
• 1994 - Nimrod project launched projects in scientific and technical
• Jul 1996 - SETI@home launched computing
• 1997 - Globus under development
• 1997 - UNICORE project launch – Major infrastructure deployments
• Mar 1997 - Condor deployed at NCSA – De facto standard technology: Globus
• 1997 - Entropia Inc founded to commercialise PC cycle scavenging ToolkitTM
• Sep 1997 - “Building a computational Grid” workshop, Argonne – Growing industrial interest
National Lab – Global Grid Forum: ~1000 people,
• Oct 1997 - SRB v1.0 released
• Jul 1998 - Foster/Kesselman: “The GRID book”
30+ countries
• Aug 1998 - Applied Meta Inc commercialises Legion – Jan 2002 - OGSA announced
• Oct 1998 - Globus v1.0.0 released – Feb 2002 - OGSA-DAI project launch
• Jun 1999 - Grid Forum 1
• Jan 2000 - UNICORE stage 2 launch – Jun 2002 - NEC Earth Simulator achieves
• Jul 2000 - SUN buys Gridware Inc  Grid Engine 35 Tflops
• Oct 2000 - NASA IPG prototype completed
• 2003
– Enterprise Attention around GRID
– GT3.0 based on OGSA released.
– Commercial Offerings from Vendors (IBM
28th Nov 2003 GRID COMPUTING
etc). 17
 In summary..
• GRID technology refers to the enabling
technology for creating a large and powerful
“VIRTUAL ORGANIZATION” out of a pool of
heterogeneous resources
• The connections are loosely coupled
• The resources are heterogeneous ranging from
computers to databases to instruments to
networks
• Non-Trivial Quality of Service offered by the VO
• Composition of VO is dynamic and flexible

28th Nov 2003 GRID COMPUTING 18

 Where GRID applies?

28th Nov 2003 GRID COMPUTING 19

 What GRID computing can do..
• Improve Resource Utilization
– Exploit underutilized resources (CPU, storage)
– Improved load balancing
• Provide high computing power
– Simulate Parallel CPU capacity
– Pool Individual computing power
• Provide additional storage
– Pool individual storage units
• Provide additional bandwidth
– Pool bandwidth from multiple units
• Enhanced collaboration among multiple stakeholders
– Beyond the enterprise
– Over a geographical spread (e.g. Collaborative research)
• Enhanced access to other resources
– Software, Licenses, Equipment
• Reliability based on Software
• IT effectiveness
– Ease of management of IT infrastructure
• Ability to execute Parallelizable applications

28th Nov 2003 GRID COMPUTING 20

 Domains where GRID can be
applied…
• Scientific Domain • Government
– GriPhyN (US Grid Physics Network for Data- – Tax Processing
intensive Science) for Elementary particle
physics, gravitational wave astronomy, optical – Census applications
astronomy (digital sky survey) – Online Processing
– DataGrid (led by CERN) for Analysis of data
from scientific exploration • Atmospheric Science
• Financial Domain – Imagery and Geo-Spatial
– Analytics (Risk Analysis and Modeling) Intelligence
– Portfolio Rebalancing – Weather /Ocean Forecasting
– Treasury and Federal Banking
• Defense and Security
• Life Sciences and Pharma
–
– Nuclear weapons advanced
Data Mining for Bioinformatics
– Drug Discovery
simulation and modeling
• – Threat Analysis
Corporate Applications
– Business Intelligence – Cryptanalysis
– IT Effectiveness – Weapons Performance Analysis
– Digital Content Distribution • Economics
• Automotive – Econometric modeling
– Collaborative Design
– Combat Systems Design
– Stealth Design of defense systems

28th Nov 2003 GRID COMPUTING 21

 A typical use case for data mining
in Bio-Informatics

Mining Search Request

Bio Data Base 1
Bio-Mining Application

Request for a Resource

data mining resource
Search Request Store Intermediate Result

Bio Data Base 2

Storage Store Intermediate Result
Request for a Resource
Transient storage
resource

28th Nov 2003 GRID COMPUTING 22

 How is GRID constructed?

28th Nov 2003 GRID COMPUTING 23

 Overall GRID Architecture
Internet
GRID
Application

Application
Collective

Resource

Connectivity Transport
Internet
Fabric Link
Source: The Anatomy of the GRID, Foster, Kesselman and Teucke

28th Nov 2003 GRID COMPUTING 24

 Fabric Layer
• Fabric layer: Provides the resources to which shared
access is mediated by Grid protocols.
• Example: computational resources, storage systems,
catalogs, network resources, and sensors.
• Fabric components implement local, resource
specific operations.
• Richer fabric functionality enables more sophisticated
sharing operations.
• Sample resources: computational resources, storage
resources, network resources, code repositories,
catalogs.

28th Nov 2003 GRID COMPUTING 25

 Connectivity Layer
• Communicating easily and securely.
• Connectivity layer defines the core
communication and authentication protocols
required for grid-specific network functions.
• This enables the exchange of data between
fabric layer resources.
• Support for this layer is drawn from TCP/IP’s IP,
TCL and DNS layers.
• Authentication solutions: single sign on, etc.

28th Nov 2003 GRID COMPUTING 26

 Resources Layer
• Resource layer defines protocols, APIs, and SDKs for
secure negotiations, initiation, monitoring control,
accounting and payment of sharing operations on
individual resources.
• Two protocols information protocol and management
protocol define this layer.
• Information protocols are used to obtain the information
about the structure and state of the resource, ex:
configuration, current load and usage policy.
• Management protocols are used to negotiate access to
the shared resource, specifying for example qos,
advanced reservation, etc.

28th Nov 2003 GRID COMPUTING 27

 Collective Layer
• Coordinating multiple resources.
• Contains protocols and services that capture
interactions among a collection of resources.
• It supports a variety of sharing behaviors without
placing new requirements on the resources
being shared.
• Sample services: directory services,
coallocation, brokering and scheduling services,
data replication service, workload management
services, collaboratory services.

28th Nov 2003 GRID COMPUTING 28

 Applications Layer
• These are user applications that operate
within VO environment.
• Applications are constructed by calling
upon services defined at any layer.
• Each of the layers are well defined using
protocols, provide access to useful
services.
• Well defined APIs also exist to work with
these services.

28th Nov 2003 GRID COMPUTING 29

 Implementations
• Till date mostly proprietary
implementations
• Globus (Open Source) Toolkit the most
popular one
• OGSA (Open Grid Services Architecture),
a move for open standards for grid
• Globus Toolkit 3.0 recently released
OGSA compliant

28th Nov 2003 GRID COMPUTING 30

 Globus Toolkit 2.0 : Proprietary
Implementation
• Grid protocols (GSI, GRAM, …) enable resource sharing within
virtual orgs; toolkit provides reference implementation ( = Globus
Toolkit services)

MDS-2 Soft state

registration;
GSI User
Reliable (Monitor./Discov. Svc.) enquiry
remote
(Grid invocation Gatekeeper Reporter GIIS: Grid
(registry + Information
Security Authenticate & (factory)
discovery)
create proxy Other GSI- Index Server
Infrastruc- Create process Register (discovery)
credential authenticated
ture)
User User remote service
process #1 process #2 requests
Proxy Other service
Proxy #2
GRAM (e.g. GridFTP)
(Grid Resource Allocation & Management)
 Protocols (and APIs) enable other tools and services
for membership, discovery, data mgmt, workflow, …
Source: Grid Services and Web Services Tutorial: GlobusWorld 2003
28th Nov 2003 GRID COMPUTING 31
 OGSA = Standards for GRID
Implementations

28th Nov 2003 GRID COMPUTING 32

Open Grid Services Architecture
• Based on Service oriented Architecture
– virtualize resources
– unify resources/services/information
• Leverages useful Web Services properties
– Standards for service description and discovery
– Leverage commercial efforts in Web Services
• Leverages existing grid systems properties
– Service Semantics
– Lifecycle management
– Reliability and Security models
– Resource Management
– Authorization etc.
• Provides a unifying architecture for computational Grids

28th Nov 2003 GRID COMPUTING 33

 Web Services
“ A web service is a software application identified by a URI, whose
interfaces and bindings are capable of being defined, described and
discovered by XML artifacts. A web service supports direct
interaction with other software agents using XML based messages
exchanged via internet based protocols ” Source: WS Architecture Working
Group W3C

 Web Services are software applications based on open

standards. These applications can be :
•Published,
•Searched,
•Located, and
•Invoked by other applications on internet/intranet/extranet
Strict adherence to standards makes it easy for one
application to talk to another
XML is the lingua-franca of communication
28th Nov 2003 GRID COMPUTING 34
 OGSA is an implementation of a Service Oriented
Architecture (SOA)
 Service Providers

 Provide service functionality which is

SERVICE published by the Service discovery agency
PROVIDER
 Discovery Agency

BI  Maintain a registry of services, their interface

ND descriptions, provider information and
H
IS

invocation methods
BL
PU

 Service Requestors

DISCOVERY SERVICE  Locate required service from the services

AGENCY REQUESTOR
FIND published by the Discovery agency, and get
all the information for binding to the service
from the agency

28th Nov 2003 GRID COMPUTING 35

 The Open Grid Services
Architecture

schemas
More specialized &

Other
domain-specific
services

OGSA schemas
OGSA services: registry,
authorization, monitoring, data
access, management, etc., etc.

Open Grid Services Infrastructure

Web Services
Host. Env. & Protocol Bindings
HostingEnvironment
Hosting Environment Transport
Protocol

28th Nov 2003 GRID COMPUTING 36

 Open Grid Services Architecture
- Detailed
Users in Problem Domain X

Applications in Problem Domain X

Application & Integration Technology for Problem Domain X
Generic Virtual Service Access and Integration Layer
Job Submission Brokering Workflow Structured Data
Registry Banking
OGSA
Authorisation Integration

Data Transport Resource Usage Transformation Structured Data Access

OGSI: Interface to Grid Infrastructure

Web Services: Basic Functionality

Compute, Data & Storage Resources

Structured Data
Distributed Relational XML Semi-structured
-

Virtual Integration Architecture

28th Nov 2003 GRID COMPUTING 37
Concepts in OGSA
• Naming
Globally & uniquely identify a grid service instance by a GSH (Grid Service Handler in the form of
a URI). <for example>
GSH is a URI
(http://192.168.0.1:8080/ogsa/services/base/multirft/MultiFileRFTFactoryService)
Can be thought of as a network pointer to a grid service but does not provide enough information to
access a grid service.
GSH needs to be resolved to a GSR in order to access a grid service instance.
GSR is A temporal, binding specific end-point that provide access to a grid service instance.
GSR is a WSDL document that describe how to reach a grid service instance

• Factories
Create new grid service instances and maintain a group of service data elements which can be
queried. A factory have a associated registry to keep track of instances and enable discovery.
• Instances
Client communicate with Grid service instance via GSR (Grid Service Reference). GSH is
mapped to the appropriate GSR via the registry.
• Stateful Web Services
A grid service instance has a state.

28th Nov 2003 GRID COMPUTING 38

 A run-time view of Open Grid Services
Infrastructure (OGSI)
Resource
allocation
Create
Service
Authentication
Grid Service
& authorization Service Service
Handle
are applied to factory requestor
all requests (e.g. user
Service data application)
Keep-alives
Notifications
Service Service
invocation discovery

Register
Service Service Service
instances registry

Interactions standardized using WSDL 39

28 Nov 2003
th
GRID COMPUTING
XML definition for a WSDL GSR
targetNamespace = http://www.gridforum.org/namespaces/2003/03/OGSI”
<xsd:complexType name="WSDLReferenceType">
<xsd:complexContent>
<xsd:extension base="ogsi:ReferenceType">
<xsd:sequence>
<xsd:element ref="wsdl:definitions"/>
</xsd:sequence>
</xsd:extension>
</xsd:complexContent>
</xsd:complexType>

XML definition for a GSH

<targetNamespace = “http://www.gridforum.org/namespaces/2003/03/OGSI”
<xsd:element name="handle" type="ogsi:HandleType"/>
<xsd:simpleType name="HandleType">
<xsd:restriction base="xsd:anyURI"/>
</xsd:simpleType>
28th Nov 2003 GRID COMPUTING 40
 WSDL: Recap Data Types
Abstract Interface
HTTP Binding

Endpoint

28th Nov 2003 GRID COMPUTING 41

 GWSDL: Differences and Example
•Differences
•Interface Inheritance
•Ability to describe additional information elements with interface definitions
•serviceData – special elements for life cycle (can be static or dynamic)
•Typical attributes: goodFrom, goodUntil, availableUntil
Example:
<wsdl:definitions>
<wsdl:types>…</wsdl:types>
<wsdl:message>…</wsdl:message>
…
<gwsdl:portType name=“foo”
extends=“ns:bar ogsi:GridService”>

<wsdl:operation name=“op1”>…</wsdl:operation>
<wsdl:operation name=“op2”>…</wsdl:operation>

<ogsi:serviceData … />

</gwsdl:portType>
…
</wsdl:definitions>

28th Nov 2003 GRID COMPUTING 42

 OGSI Grid Service Behavior portTypes
Service Data
Interface Description Default service data
PortType Name elements defined
and Operations (static) values
by this portType
All Grid services
implements this <ogsi:
interface and Interfaces findServiceDataExten
provides these serviceDataName sibility
operations and factoryLocator inputElement="queryB
behaviors. GridServiceHandle yServiceDataNames" /
Operations GridServiceRefrence >
GridService (required)
findServiceData findServiceDataExten <ogsi:
setServiceData sibility setServiceDataExtens
requestTermination setServiceDataExtens ibility
TimeAfter ibility inputElement="delete
requestTermination terminationTime ByServiceDataNames
TimeBefore " />
destroy
To create a new Grid
service.
Factory (optional) 1.createServiceExtensibility None
Operations
1.createService
A service provided
mechanism to resolve
HandleResolver (optional) a GSH to a GSR handleResolverScheme None
Operations
1.FindByHandle

28th Nov 2003 GRID COMPUTING 43

 GridService portType
Operations

Operation Description

<TB>findServiceData Query information about the Grid

service instance
setServiceData Modify service data values
requestTerminationAfter Specify earliest desired
termination time
requestTerminationBefore Specify latest desired
termination time
destroy Terminate Grid service instance

28th Nov 2003 GRID COMPUTING 44

 OGSI Grid Service Notification portTypes
Service Data
Interface Description Default service data
PortType Name elements defined
and Operations (static) values
by this portType
This enables a client to
subscribe for <ogsi:
notification based on a subscribeExtensibility
notifiableServiceDataName
NotificationSource (optional) service data value inputElement="subscri
subscribeExtensibility
change. beByServiceDataNam
Operations es" />
1.subscribe
Implementing this interface
enables a Grid service
instance to receive
notification messages
NotificationSink (optional) None None
based on a
subscription.
Operations
1.deliverNotification
Calling a subscription of a
Notification Source
results in the creation
NotificationSubscription subscriptionExpression
of a subscription Grid None
(optional) sinkLocator
service.
Operations
None defined

28th Nov 2003 GRID COMPUTING 45

 OGSI Grid Service Grouping
Behavior
Service Data
Interface Description Default service data
PortType Name elements defined
and Operations (static) values
by this portType
An abstract interface to represent a
grouping of zero or more
services. This interface
extends the GridService
MembershipContentRule
ServiceGroup (optional) portType. None
entry
Operations
None defined but can use
operations defined in
GridService portType.

This interface extends the

ServiceGroup interface and
provides operations to
<ogsi: removeExtensibility
manage a ServiceGroup
addExtensibility inputElement=
ServiceGroupRegistration (optional) including adding/delete a
removeExtensibility "matchByLocatorEquivalence
service to/from a group.
" />
Operations
add
remove

This is a representation of an
individual entry of a
ServiceGroup and is created
on ServiceGroupRegistration
"add". Each entry contains a
service locator to a member memberServiceLocator
ServiceGroupEntry (optional)  
Grid service and information content
about the member service as
defined by the Service group
membership rule (content).
Operations
None defined
28th Nov 2003 GRID COMPUTING 46
 Example:
Use Case Revisited: Database Service for BioInformatics
• A DBaccess Grid service will support at least two
portTypes Grid
Service DBaccess
– GridService
– DBaccess
Name, lifetime, etc.
• Each has service data
DB info
– GridService: basic introspection information, lifetime, …
– DBaccess: database type, query languages supported,
current load, …, …
• Maybe other portTypes as well
– E.g., NotificationSource

Source: Grid Services and Web Services Tutorial: GlobusWorld 2003

28th Nov 2003 GRID COMPUTING 47
 Transient Database Service
“What services “Create a database
can you create?” service”

Grid Grid
DBaccess
Service Service
Factory DBaccess

Instance name, etc. Name, lifetime, etc.

“What database Factory info DB info

services exist?”

Grid Grid
Service Registration Service DBaccess

Instance name, etc. Name, lifetime, etc.

Registry info DB info

Source: Grid Services and Web Services Tutorial: GlobusWorld 2003

28th Nov 2003 GRID COMPUTING 48
 Example:
Data Mining for Bioinformatics
Community
Registry Mining
Factory Database
Service

BioDB 1

User Compute Service Provider

.
Application .
.
.
.
.
“I want to create
Database
a personal database Database Service
containing data on Factory
e.coli metabolism”
BioDB n

Storage Service Provider

Source: Grid Services and Web Services Tutorial: GlobusWorld 2003

28th Nov 2003 GRID COMPUTING 49
 Example:
Data Mining for Bioinformatics
“Find me a data Community
mining service, and Registry Mining
Factory Database
somewhere to store Service
data”
BioDB 1

User Compute Service Provider

.
Application .
.
.
.
.
Database
Database Service
Factory

BioDB n

Storage Service Provider

Source: Grid Services and Web Services Tutorial: GlobusWorld 2003

28th Nov 2003 GRID COMPUTING 50
 Example:
Data Mining for Bioinformatics
Community
Registry Mining
GSHs for Mining Factory Database
Service
and Database
factories
BioDB 1

User Compute Service Provider

.
Application .
.
.
.
.
Database
Database Service
Factory

BioDB n

Storage Service Provider

Source: Grid Services and Web Services Tutorial: GlobusWorld 2003

28th Nov 2003 GRID COMPUTING 51
 Example:
Data Mining for Bioinformatics
Community
Registry Mining
Factory Database
“Create a data mining Service
service with initial
lifetime 10” BioDB 1

User Compute Service Provider

.
Application .
.
.
.
.
“Create a Database
database with initial Database Service
lifetime 1000” Factory

BioDB n

Storage Service Provider

Source: Grid Services and Web Services Tutorial: GlobusWorld 2003

28th Nov 2003 GRID COMPUTING 52
 Example:
Data Mining for Bioinformatics
Community
Registry Mining
Factory Database
“Create a data mining Service
service with initial
lifetime 10” Miner BioDB 1

User Compute Service Provider

.
Application .
.
.
.
.
“Create a Database
database with initial Database Service
lifetime 1000” Factory

BioDB n
Database

Storage Service Provider

Source: Grid Services and Web Services Tutorial: GlobusWorld 2003

28th Nov 2003 GRID COMPUTING 53
 Example:
Data Mining for Bioinformatics
Community
Registry Mining
Factory Database
Query Service

Miner BioDB 1

User Compute Service Provider

.
Application .
.
.
Query .
.
Database
Database Service
Factory

BioDB n
Database

Storage Service Provider

Source: Grid Services and Web Services Tutorial: GlobusWorld 2003

28th Nov 2003 GRID COMPUTING 54
 Example:
Data Mining for Bioinformatics
Community
Registry Mining
Factory Database
Query Service

Miner BioDB 1
Keepalive
User Compute Service Provider
.
Application .
.
.
Query .
.
Database
Database Service
Keepalive Factory

BioDB n
Database

Storage Service Provider

Source: Grid Services and Web Services Tutorial: GlobusWorld 2003

28th Nov 2003 GRID COMPUTING 55
 Example:
Data Mining for Bioinformatics
Community
Registry Mining
Factory Database
Service

Miner BioDB 1
Keepalive
User Compute Service Provider
.
Application . Results .
.
.
.
Database
Database Service
Keepalive Factory
Results
BioDB n
Database

Storage Service Provider

Source: Grid Services and Web Services Tutorial: GlobusWorld 2003

28th Nov 2003 GRID COMPUTING 56
 Example:
Data Mining for Bioinformatics
Community
Registry Mining
Factory Database
Service

Miner BioDB 1

User Compute Service Provider

.
Application .
.
.
.
.
Database
Database Service
Keepalive Factory

BioDB n
Database

Storage Service Provider

Source: Grid Services and Web Services Tutorial: GlobusWorld 2003

28th Nov 2003 GRID COMPUTING 57
 Example:
Data Mining for Bioinformatics
Community
Registry Mining
Factory Database
Service

BioDB 1

User Compute Service Provider

.
Application .
.
.
.
.
Database
Database Service
Keepalive Factory

BioDB n
Database

Storage Service Provider

Source: Grid Services and Web Services Tutorial: GlobusWorld 2003

28th Nov 2003 GRID COMPUTING 58
 Conclusions
and
Call For Action

28th Nov 2003 GRID COMPUTING 59

 Conclusions
• GRID is based on idea of virtualization of a pool of
heterogeneous resources into one scalable virtual
organization
• GRID provides enhanced throughput, resource
utilization, non-trivial QoS, leveraging heterogeneous
resources
• GRID is suitable for computationally intensive and other
resource intensive works across multiple verticals
• GRID computing is REAL today and businesses are
leveraging GRID
• OGSA, the standard for GRID, based on web services
will be mainstream and will enhance penetration of GRID

28th Nov 2003 GRID COMPUTING 60

 Call for Action
• Read about GRID..
• Download GLOBUS Toolkit 3.0..
• Identify potential application areas of Pilot
for GRID
• Run Pilot GRIDs with a dedicated set of
services for focused problems
• Build Solutions and Take to Clients..
• GRID COMPUTING IS REAL TODAY…

28th Nov 2003 GRID COMPUTING 61

 THANKS

28th Nov 2003 GRID COMPUTING 62