Developing distributed apps? Think "JavaSpaces"
We know that Distributed Computing is all about building network-based applications as a set of processes that are distributed across a network of computing nodes that work together to solve a problem. The advantages of this approach is like performance improvement, resource sharing, fault tolerance, scalability, etc. But still, to achieve this, special care should have to be taken while design and implementation of such systems. There are many challenges in distributed computing like synchronization of process across heterogeneous networks, network latency, partial failures of processes and language incompatibilities.
JavaSpaces technology, to the rescue...
1. JavaSpaces is both, distributed programming model and an API.
2. It's a high-level coordination tool for gluing processes together into a distributed application.
3. JavaSpaces are best considered as distributed shared memory with additional features which provide transactional integrity and support for the handling of failure.
4. In JavaSpaces applications, processes don't communicate directly, but instead coordinate their activities by exchanging objects through a space, or shared memory.
5. A process can write new objects into a space, take objects from a space, or read (make a copy of) objects in a space as shown in figure below:
6. JavaSpaces uses the most common software pattern, the Master-Worker pattern. The Master hands out units of work to the 'space', and these are read, processed and written back to the space by the workers.
7. JavaSpaces provides a simplified approach to dynamic communication, coordination and sharing of objects between network resources.
Some useful pointers to know more about JavaSpaces:
Link1, Link2, Link3
