How Facebook Scales with Open Source

ReadWriteWeb has a piece on how Facebook scales to accomidate their 500 million users by using open source. The article itself is relevant to NoSQL in a couple of ways. First it's a known fact that Facebook uses multiple NoSQL databases, and let's not forget they actually created one as well.

Click to read more ...

Thrift: Scalable Cross-Language Services Implementation

Often times there are difficult decisions to be made when developing multiple applications within an organization that are supposed to integrate with each other. Do you stick with a single language making it easier to recruit talent? What if the language is ill suited for the task at hand? On the other hand do you take the risk of implementing the applications in multiple languages? While the tasks may be better fits, you now have a more diverse knowledge base that must be shared to understand how everything fits together.

Well the folks at Facebook have made it clear what their decision is:

By and large, Facebook’s engineering culture has tended towards choosing the best tools and implementations available over standardizing on any one programming language and begrudgingly accepting its inherent limitations.

The decision has made communication between various network services a critical priority. So how does Facebook solve this problem? The Thrift protocol:

a language neutral software stack implemented across numerous programming languages and an associated code generation engine that transforms a simple interface and data definition language into client and server remote procedure call libraries.

So what does this have to do with NoSQL? Thrift serves as the protocol layer for Cassandra. Much like JSON is used for other NoSQL technologies. In the paper Mark Slee, Aditya Agarwal and Mark Kwiatkowski describe Thrift types, protocol, versioning, processors and implementation details.

At some point expect us to dive into the details of the paper in more depth as there is a lot to discuss.

Read More: Thrift: Scalable Cross-Language Services Implementation

Installing and using Cassandra in just five steps

Ronald Mathies of Sodeso.nl has written an extremely nice tutorial about how to both install and use Cassandra. This series of blog postings walks the users through installation of Cassandra on both Linux and Windows, describes Cassandra's data model (which takes a bit to get your head around) and discusses details about the Thrift client.

Here are the links for the five part series:

Installing and using Apache Cassandra With Java Part 1 (Installation)
Installing and using Apache Cassandra With Java Part 2 (Data model)
Installing and using Apache Cassandra With Java Part 3 (Data model 2)
Installing and using Apache Cassandra With Java Part 4 (Thrift Client)
Installing and using Apache Cassandra With Java Part 5 (Thrift Client 2)