Win a copy of The Way of the Web Tester: A Beginner's Guide to Automating Tests this week in the Testing forum!
  • Post Reply
  • Bookmark Topic Watch Topic
  • New Topic

Developing a Client Server Application using Jini

megha malik
Posts: 1
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
I am trying to develop a client server application using Jini.
My architecture is quite straight forward..
After a review of the application services provided by and the architectures behind the various computational grids existing, it was realized that a dispatcher-based approach to cater for the dynamic load balancing, would suit ideally for our needs to this project.
In this approach, we would centralize request scheduling and completely control client-request routing. One of the network components of the Web-server system could act as a dispatcher. The dispatcher uniquely identifies each server in the system through a private address that can be at different protocol levels, depending on the architecture.
Dispatcher-based Approach
Further enhancing this approach, a centralized dispatcher receives all incoming requests, and distributes them among the Web-server nodes. The dispatcher redirects a request by specifying the appropriate status code in the response, indicating in its header the server address where the client can access the desired application.
This approach can be further implemented using two techniques:
a)Server-state-based dispatching � Since the dispatcher must be aware of the server load, each server periodically reports the number of processes in its run queue and the number of received requests per second. The dispatcher then selects the least-loaded server.
b)Location-based dispatching � client requests are redirected to the server that is evaluated as most suitable, based on its proximity to the desired server node, for reach request at a certain time.
This approach is hampered by a single-decision entity, which can be a bottleneck, with increased requests, and in centralized scheduling; a failed dispatcher could disable the system. The dispatcher however, helps achieve, fine-grained load balancing for the system under consideration.
In essence, this project could be considered as the formation of a computational grid, where there is aggregation of resources, from a heterogeneous data pool of geographically dispersed compute resources linked by a network.
We thus make use of a middleware that provides a programming abstraction as well as masking the heterogeneity of the underlying network, hardware and operating systems.
To understand the architecture further, we could analyze it as a consumer-producer model.
The Consumer
-Provides an interface to the clients.
-It is a launch point for applications and a collection point for results
-It is here that the registration of clients, task submissions and result retrievals are carried out.
The Producer
-Provides the computing power
-Executes tasks and is also the work engine of the system
-The producer, which may also be known as the Web-server node, may also hold a local scheduler.
The Resource Broker
-aka the Dispatcher, as discussed above.
-Provides for authentication
-It is the center of the application execution control
-Handles resource management, such as scheduling and load balancing
-Holds the task pool or the request queue.
Thus, discussed above is a brief overview of the kind of architecture that could be implemented to serve the purpose of this project.
Here the broker could be implemented using Jini.
I need help in implementing this project.
Could someone me in tellign me and how where to start.. and where ot go from there...
I am running short on time.. and need to act fast...
Thanks a lot
  • Post Reply
  • Bookmark Topic Watch Topic
  • New Topic