sábado, 19 de abril de 2014

Introduction to Apache Cassandra - Summary

Hi devs!

As we announced on the last entry, on April 1st, we attended a terrific Cassandra talk with Colin Clark as the speaker, on which he explained the use of Cassandra they made at Datastax for a distributed concurrent and high availability Stock Exchange monitoring and management application (a very simplified view of what the application actually did, of course).

At first, we were given an overview of the project, what it entailed, what were the client's needs and the protocols used for information exchange. With that information, we were shown how the system was built around those foundations. Colin described the different layers of their system, the ever functioning monitoring and processing layer that aggregated all real time data coming from the stock exchange sources so as to empower business analysts to take sound strategic decisions.

This information processing system couldn't have been made without Cassandra, the key technology, since this NOSQL database is designed from the ground up to be scalable, thanks to it's distributed node architecture. As was debated later on the questions round, the alternatives weren't up to the needs of the application, in performance, availability and cost (both in time to implement and on licenses to pay).

Colin Clark also gave us some pointers on best (and worst) practices to migrate a relational database backed application to a nosql like Cassandra.

All in all, a very good talk, and one that will open further collaborations with Datastax. Stay tuned for more info on that the next months, big things are yet to come this year!

Talk's video:

No hay comentarios:

Publicar un comentario en la entrada