We are writing a book on large-scale graph processing with Apache Giraph. Giraph is being currently used at Facebook to run graph analytics on their graph with a trillion relationships across hundreds of machines within minutes [1]. The book covers from technical challenges about designing and building a processing engine that runs iterative graph algorithms efficiently, how to develop scalable algorithms for graph analytics, to the practical API of Apache Giraph. We thought you guys might be interested in these use cases for your own work, and hopefully find the content useful to you. We hope to get some useful feedback on how to improve the book to make it useful to the broadest audience.Do you have use-cases where you think Giraph would fit in your application architecture? In other words, what would you think would be useful to discuss in the book?
You can find the book at http://www.manning.com/martella/ with the free first chapter.
[1] https://www.facebook.com/notes/facebook-engineering/scaling-apache-giraph-to-a-trillion-edges/10151617006153920