This post describes a prototype implementation of a simple PAAS built on the Hadoop YARN framework and the key findings from the experiment. While there are some advantages to using Hadoop YARN, there is at least one unsolved issue that would be difficult to overcome at this point.
Hadoop is so popular these days as Big Data is one of the major hot topics people are intererested in. Similarly, PaaS (Platform as a Service) is also popular as Cloud Computing is one of the hot topics. Then naturally one question came to us: can we combine both Hadoop and PaaS to satisfy the two hot topics – Big Data and Cloud? At the same time, Hadoop YARN (i.e. MapReduce2 or MR2) architecture became much more flexible compared to the previous version that the idea seemed to become more real.
I went ahead and implemented a proof of concept with Hadoop Yarn. Here I’d like to share its architecture and interesting findings from it.