Distributed Computing Paradigms L2

8/8/2019 Distributed Computing Paradigms L2

1/26

Distributed ComputingParadigms

Lecture-2


2/26

Distributed App. Paradigms

high

object space

network services, object request broker, mobile agent

remote procedure call , remote method invocation

client-server

message passing

low

IT636 - Distributed Computing: Fundamentals and Applications


3/26

The Message Passing Paradigm

Message passing is the most fundamental paradigm for distributed

a lications.

A process sends a message representing a request. The message is delivered to a receiver, which processes the request,

and sends a message in response.

In turn, the reply may trigger a further request, which leads to asubsequent reply, and so forth. -

Process AProcess B

a message


Message passing


4/26

The Message Passing Paradigm - 2

The basic operations required to support the,

receive.

For connection-oriented communication, theoperations connect and disconnect are also

required.



5/26

The Client-Server Paradigm

Perhaps the best known paradigm for network applications,the client-server model assi ns as mmetric roles to two

collaborating processes. One process, the server, plays the role of a service provider

. ,the client, issues specific requests to the server and awaits itsresponse.

service requesta client process

a server process

a serviceServer host

Client host

...


The Client-Server Paradigm, conceptual


6/26

The Client-Server Paradigm - 2

Simple in concept, the client-server model provides anefficient abstraction for the deliver of network services.

Operations required include those for a server process tolisten and to accept requests, and for a client process to issue

.

By assigning asymmetric roles to the two sides, event

synchronization is simplified: the server process waits for, .

Many Internet services are client-server applications. Theseservices are often known by the protocol that the application

. ,FTP, DNS, finger, gopher, etc.



7/26

The Peer-to-Peer Architecture

In system architecture and networks, peer-to-peer

services are direct exchanged between computersystems.

exchange of information, processing cycles, cache

storage, and disk storage for files.. n suc an arc tecture, computers t at ave

traditionally been used solely as clientscommunicate directly among themselves and can

act as both clients and servers, assuming whateverrole is most efficient for the network.



8/26

The Peer-to-Peer Distributed Computing

In the peer-to-peer paradigm, the participating processes playe ual roles with e uivalent ca abilities and res onsibilities

(hence the term peer). Each participant may issue arequest to another participant and receive a response.

process 1

request

response

request

response

process 2



9/26

Peer-to-Peer distributed computing

Whereas the client-server paradigm is an ideal model for acentralized network service the eer-to- eer aradi m is

more appropriate for applications such as instant messaging,peer-to-peer file transfers, video conferencing, andcollaborative work. It is also possible for an application to bebased on both the client-server model and the peer-to-peermodel.

A well-known example of a peer-to-peer file transfer service isNapster.comor similar sites which allow files (primarily audiofiles) to be transmitted among computers on the Internet. It

makes use of a server for directory in addition to the peer-to-peer computing.



10/26

Peer-to-Peer distributed computing

The peer-to-peer paradigm can be implemented with facilitiesusing any tool that provide message-passing, or with ahigher-level tool such as one that supports the point-to-point

model of the Message System paradigm.

For web applications, the web agentis a protocol promoted bye e u c rus rgan za on or peer-

to-peer interprocess communication

Project JXTA is a set of open, generalized peer-to-peer,

PDA, PC to server) on the network to communicate andcollaborate. JXTA is short for Juxtapose, as in side by side. Itis a recognition that peer to peer is juxtapose to client server

or Web based computing -- what is considered today'straditional computing model.



11/26

The Message System Paradigm

The Message System or Message-Oriented Middleware (MOM)

paradigm is an elaboration of the basic message-passing paradigm.

In this paradigm, a message system serves as an intermediaryamong separate, independent processes.

The message system acts as a switch for messages, through which

processes exchange messages asynchronously, in a decoupled

manner.

A sender deposits a message with the message system, whichorwar s o a message queue assoc a e w eac rece ver.

Once a message is sent, the sender is free to move on to other

tasks.receivers

...message system sender


...


12/26

Two subtypes of Message models

The Point-To-Point Message Model

,

the sender to the receivers message queue. Unlike thebasic message passing model, the middleware provides a

,receiving to be decoupled. Via the middleware, a senderdeposits a message in the message queue of the receivingrocess. A receiving rocess extracts the messages from its

message queue, and handles each one accordingly.

Compared to the basic message-passing model, thisaradi m rovides the additional abstraction for

asynchronous operations. To achieve the same effect withbasic message-passing, a developer will have to make use ofthreads or child processes.



13/26

Publish/Subscribe Message Model

In this model, each message is associated with a specificto ic or event. A lications interested in he occurrence

of a specific event may subscribe to messages for thatevent. When the awaited event occurs, the processpublishes a message announcing the event or topic. Themiddleware message system distributes the message toall its subscribers.

The ublish/subscribe message model offers a owerfulabstraction for multicasting or group communication.The publishoperation allows a process to multicast to agroup of processes, and the subscribeoperation allows

a process to listen for such multicast.



14/26

Toolkits for the Message-System

The MOM paradigm has had a long history in.

Message Queue Services (MQS) have been inuse since the 1980s.

The IBM MQ*Series is an example of such a

facility. Other existing support for this paradigm are

Microsofts Message Queue (MSQ),

Javas Messa e Service



15/26

Remote Procedure Call

As applications grew increasingly complex, it becamedesirable to have a paradigm which allows distributedsoftware to be programmed in a manner similar to

conventional applications which run on a singleprocessor.

e emo e roce ure a mo e prov es sucan abstraction. Using this model, interprocess

communications proceed as procedure, or function,calls which are familiar to a lication ro rammers.

A remote procedure call involves two independentprocesses, which may reside on separate machines. Aprocess, A, wishing to make a request to another

process, B, issues a procedure call to B, passing with thecall a list of argument values. As in the case of localprocedure calls, a remote procedure call triggers a


At the completion of the procedure, process Breturns a

value to process A.


16/26

Remote Procedure Call - 2

Process A Process B

proc1(arg1, arg2)

proc arg

proc arg ,arg ,arg



17/26

Remote Procedure Call - 3

RPC allows programmers to build network applicationsusing a programming construct similar to the localprocedure call, providing a convenient abstraction for

both interprocess communication and eventsynchronization.

nce s n ro uc on n e ear y s, e emo eProcedure Call model has been widely in use in networkapplications.

Calls.

The Open Network Computing Remote Procedure Call,evolved from the RPC API ori inated from Sun

Microsystems in the early 1980s. The Open Group Distributed Computing Environment

(DCE) RPC.


Both APIs provide a tool, rpcgen, for transforming

remote procedure calls to local procedure calls to thestub.


18/26

The Distributed Objects Paradigms

The idea of applying object orientation to distributeda lications is a natural extension of ob ect-oriented

software development. Applications access objects distributed over a network.

,which an application obtains access to services.

Object-oriented paradigms include: Remote method invocation (RMI)

Network services

Ob ect re uest broker

Object spaces



19/26

Remote Method Invocation (RMI)

Remote method invocation is the object-oriented.

In this model, a process invokes the methods in anob ect, which ma reside in a remote host.

As with RPC, arguments may be passed with the

invocation.

method1

Process 1

Process 2

remote method invocation

method2

a remote object


The Remote Method Call Paradigm


20/26

The Network Services Paradigm

In this paradigm, service providers register themselves withdirector servers on a network. A rocess desirin aparticular service contacts the directory server at run time,and, if the service is available, will be provided a reference tothe service. Using the reference, the process interacts withthe service.

This paradigm is essentially an extension of the remote

method call aradigm. The difference is that service objectsare registered with a global directory service, allowing them tobe look up and accessed by service requestors on afederated network. Directory service

Javas Jini

service object

IT636 - Distributed Computing: Fundamentals and ApplicationsService requestor


21/26

Object Request broker Paradigm

In the object broker paradigm , an application issuesre uests to an ob ect re uest broker ORB whichdirects the request to an appropriate object that providesthe desired service.

invocation model in its support for remote object access.The difference is that the object request broker in this

aradigm functions as a middleware which allows anapplication, as an object requestor, to potentially accessmultiple remote (or local) objects.

The re uest broker ma also function as an mediator for

heterogeneous objects, allowing interactions amongobjects implemented using different APIs and /or runningon different platforms. Ob ect


Requestorec

Object Request Broker


22/26

Object Request broker Paradigm - 2

This paradigm is the basis of the Object

Request Broker Architecture) architecture. Tool kits based on the architecture include:

Inprises Visibroker

Javas Interface Development Language (Java IDL) , ,

Inc.



23/26

The Object Space Paradigm

Perhaps the most abstract of the object-,

assumes the existence of logical entities knownas object spaces.

The participants of an application converge in acommon object space.

object space, and requesters who subscribe tothe space access the entries. providerrequestor

requestor

read

writeread

IT636 - Distributed Computing: Fundamentals and ApplicationsAn Object Space


24/26

The Object Space Paradigm - 2

paradigms, the object space paradigm provides a virtualspace or meeting room among provides and requesters

the detail involved in resource or object lookup neededin paradigms such as remote method invocation, object

re uest broker, or network services.

Current facilities based on this paradigm includeJavaSpaces



25/26

Component-based Technologies

Component-based technologies such as , , ,

and Enterprise Java Bean are also based ondistributed-object paradigms, as componentsare essentially specialized, packaged objectsdesigned to interact with each other through

standardized interfaces.

In addition, application servers, popular forenterprise applications, are middleware

ac es w c prov e access o o ec s orcomponents.



26/26

The Mobile Agent Paradigm

A mobile agent is a transportable program or object.

,

host. The agent travels from host to host according to an

.

At each stop, the agent accesses the necessary

resources or services, and performs the necessary tasks.Host 2

Host 1

Host 3

agent


Host 4

Documents

Distributed Computing Paradigms L2