Running Production Hadoop Clusters in Docker Containers

We are very happy to have Nasser Manesh to talk about deployment and production side of the big data analysis. An important part that many have to deal with. here are the talk details.

Title: Running Production Hadoop Clusters in Docker Containers

by Nasser Manesh

Summary:

In this talk I will share my experience and "what works and what doesn't" lessons, from an operations point of view, on running production Hadoop clusters in Docker containers. Docker has become a popular platform among developers, but the number of installations that use Docker in production is still limited. Combine that with the specific requirements of Hadoop as a distributed system, and you will find out that combining Docker and Hadoop is not trivial. The focus of the talk will be on containers, why using them in a production environment, different models in which Docker and Hadoop can be used together, lessons learned as we experimented with Docker at Altiscale, and how we are currently using Docker for our Hadoop clusters.

Bio:

Nasser Manesh has 25 years of Unix, infrastructure, distributed systems, and operations experience. He has founded startups and has been in CTO, VP Engineering, infrastructure architect, and SRE roles. He is currently focused on virtualization, cluster orchestration, and automation of Big Data infrastructure at Altiscale, the leading provider of Hadoop as a service in the cloud.

Agenda:

Door opens: 6:30 pm

Social time: 6:30 pm - 7:00 pm

talk : 7:00 -- 8:15 pm

Q & A : 8:15 - 8:30 pm

individual Q&A : 8:30 -- 8:45 pm

office close : 9 pm