sai deepthi yarlagadda study of java based grid computing april 30th 2003

15
SAI DEEPTHI YARLAGADDA STUDY OF JAVA BASED GRID COMPUTING APRIL 30TH 2003

Post on 19-Dec-2015

217 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: SAI DEEPTHI YARLAGADDA STUDY OF JAVA BASED GRID COMPUTING APRIL 30TH 2003

SAI DEEPTHI YARLAGADDA

STUDY OF JAVA BASED GRID COMPUTING

APRIL 30TH 2003

Page 2: SAI DEEPTHI YARLAGADDA STUDY OF JAVA BASED GRID COMPUTING APRIL 30TH 2003

WHAT IS GRID COMPUTING

• A grid is a collection of distributed computing resources available over a local or wide area network that appear to end user or application as one large virtual computing system

Page 3: SAI DEEPTHI YARLAGADDA STUDY OF JAVA BASED GRID COMPUTING APRIL 30TH 2003

TYPES OF GRIDS

• Cluster grid:Consists of one or more systems working together to provide a single point of access to users.

• Campus grid:Multiple projects or departments share computing resources in a cooperative way

• Global grid:The power of distributed resources to anywhere in the world

Page 4: SAI DEEPTHI YARLAGADDA STUDY OF JAVA BASED GRID COMPUTING APRIL 30TH 2003

JAVA COMMODITY GRID KIT

• A COMMODITY GRID TOOLKIT (CoG Kit) defines and implements a set of general components that map grid functionality into a commodity environment/framework.

(ex:java CoG Kit,CORBA CoG Kit,DCOM CoG Kit ETC)• Why use Java CoG Kit, the ten reasons to use java CoG Kit are The Language The Maintenance The Class Library The Performance The Components The Gadgets The Deployment The Industry The Portability The Community

Page 5: SAI DEEPTHI YARLAGADDA STUDY OF JAVA BASED GRID COMPUTING APRIL 30TH 2003

RELATED WORKS

Page 6: SAI DEEPTHI YARLAGADDA STUDY OF JAVA BASED GRID COMPUTING APRIL 30TH 2003

ALiCE - A JAVA BASED GRID COMPUTING SYTEM

ALiCE is a Adaptive and scaLable Internet-based Computing engine

• Supports computing on demand and resource sharing.

• Supports the development and execution of generic grid applications.

• It is a middleware that provides a programming abstraction as well as masking the heterogeneity of the underlying networks ,hardware and operating systems.

Page 7: SAI DEEPTHI YARLAGADDA STUDY OF JAVA BASED GRID COMPUTING APRIL 30TH 2003

MAIN FEATURES

• Job parallelism to maximize throughput and object parallelism to maximize performance

• Distributed load balancing algorithm application driven distributed load balancing ; task replications for fault-tolerant and to meet

performance deadline• Levels of security at varying costs• Usage of Java,Java Jini/Java Spaces technologies• Adaptive parallelism,cross-platform

portability,extensibility and scalability

Page 8: SAI DEEPTHI YARLAGADDA STUDY OF JAVA BASED GRID COMPUTING APRIL 30TH 2003

ALiCE PRODUCER CONSUMER MODEL

Consumers(clients):

Interface to users

Launch point for applications

Collection point for results

Producers(servers):

Provide computing power

Executes tasks

Resource brokers(RB):

Authentication

Application execution control

Resource management

Page 9: SAI DEEPTHI YARLAGADDA STUDY OF JAVA BASED GRID COMPUTING APRIL 30TH 2003

ALiCE EXECUTION MODEL

Producer 2

Producer 1

Producer n

Resource BrokerConsumer

Visualizer 1.Task generator at Resource brokerinitiates applications and produces a poolof tasks

results

2.the task execute method is run at the producers

3.Result objects produced are returned to the resource broker

4.Result collector results

Page 10: SAI DEEPTHI YARLAGADDA STUDY OF JAVA BASED GRID COMPUTING APRIL 30TH 2003

ALiCE PROGRAMMING TEMPLATES

• Task generator:Allows application to be invoked at the resource broker by invocation of the users main method

• Result collector:Allows applications to be invoked at the consumer node,waiting for results to be returned from the resource broker

• Task:Allows the producer nodes to result a return object to the resource broker upon completing the execution

• Result:Provides an interface for producer to instantiate and returns any evaluated or intermediate data

Allows users to store the result of execution

Page 11: SAI DEEPTHI YARLAGADDA STUDY OF JAVA BASED GRID COMPUTING APRIL 30TH 2003

AliCE APPLICATIONS

• Ray tracing , Mandelbrot set , georectification of satellite images

Page 12: SAI DEEPTHI YARLAGADDA STUDY OF JAVA BASED GRID COMPUTING APRIL 30TH 2003

What does grid security include

• Remote access security management preventing user forgery,server forgery,data wiretapping

and remote attack.• User rights security management adding restricted delegation on user,preventing user abuse

resources.• Task security management protecting communication between processes,system

integrity and preventing malicious programs.

Page 13: SAI DEEPTHI YARLAGADDA STUDY OF JAVA BASED GRID COMPUTING APRIL 30TH 2003

GRID APPLICATION AREAS

• Higher education

• Agricultural chemicals

• Life sciences

• Electronic industry

• Aerospace

• Government(information access)

• Automotive

• Financial services

• Workshop

Page 14: SAI DEEPTHI YARLAGADDA STUDY OF JAVA BASED GRID COMPUTING APRIL 30TH 2003

CUSTOMER EXAMPLES

• University of Pennsylvania,University of Chicago, University of California

• NASA

• UK national grid

• Butterfly.net

• Netherlands computing grid

• Distributed terascale facility

• US department of energy

• National digital mammographic archive(NDMA)

Page 15: SAI DEEPTHI YARLAGADDA STUDY OF JAVA BASED GRID COMPUTING APRIL 30TH 2003

QUESTIONS ?