Scalable Parallel and Distributed Environment (SPADE)

Today, at around 6.30 PM i went to Nivas's house, to discuss generally about SPADE (his final year project) and sample search engine, built with it. I actually learned a lot from that, and this post is to reflect the same in words. This is basically for my own purpose, just in case. If this was useful to you by any means, please leave a comment, or say your thanks to him.

SPADE -3 Properties:

1. Object Oriented Support - full support for all data types
2. Distributed cache vs DB
3. Scalability on the fly (as opposed to the need to rewrite source, recompile and redeploy in MPI)
4. Dynamic scheduling - No barrier synchronization - no static binding of code to machine
5. Asynchronous Communication as opposed to blocking sends and receives
6. Fault tolerance - no single point of failure

SPADE Abstract can be found here: http://bit.ly/8XMeTr

I'm currently planning to work with him, on this environment to implement the Research Engine, for ThinkQuest 2011. Again, time'll answer everything. I'll keep this updated, as we progress further.