1: Couchbase and Hadoop. 2: Sub-millisecond response times with Couchbase.

Connecting Couchbase to Hadoop: Couchdoop

We needed a bridge between the real-time tier, where we used Couchbase, and the batch tier, built on Hadoop. When we couldn’t find a suitable option, we built our own: Couchdoop – an open-source Hadoop connector for Couchbase.

Based on our experience with Couchdoop, we will discuss best practices in creating connectors for Hadoop and NoSQL DBs. We’ll address the challenges we encountered while developing Couchdoop and share how we tuned it for performance.

Together with Bigstep, we will also show how much throughput that can be squeezed from a Hadoop connector. We have benchmarked Couchdoop for performance and we’ll talk about the behavior you can expect and tweaks that can improve the performance of your big data setup.

Speakers: Calin Burloiu (Big Data Engineer at Avira) and Radu Pastia (Big Data Team Leader at Avira)

Achieving sub-millisecond response time with Couchbase

We set out to see just what Couchbase can do and ran a series of performance benchmarks, looking to achieve the infamous sub-millisecond response time. We not only made it, but found a number of very interesting things along the way. For instance, that there’s a difference between GET/PUT operations and QUERY operations when multiple instances are added to the cluster. After weighing all the relevant factors, we started looking for the hardware setup that would provide the best performane/price in Couchbase deployments.

During the presentation we will share all of these findings, as well as a series of best practices in optimizing Couchbase for performance that will work on most infrastructures.

Speaker: Alex Bordei (Product Manager at Bigstep)