OpenStack Presentation Voting

Thank you for all of your input!

Voting for presentations at the OpenStack Paris Summit has closed.

"Targeting OpenStack Clouds"

OpenStack Data Processing (Sahara): Use cases, direction and performance

The Sahara project (ex. Savanna), an integrated project in Juno under the OpenStack Data Processing program, provides users the ability to provision and manage Hadoop clusters on OpenStack, and has seen a great deal of progress, development, and changes during the Juno development cycle. The focus of the project is on two primary use cases: on-demand cluster provisioning and on-demand Hadoop task execution (Elastic Data Processing).

In this session we will discuss Sahara use cases and look at the value of working with Hadoop / Big Data on OpenStack using Sahara. We’ll provide an overview of the possible directions for integration between OpenStack and data processing tools and frameworks, such as:

  • Using OpenStack Keystone as the Identity provider for Apache Hadoop
  • Using Oozie for securely running clusters and workloads on top of those clusters

These solutions make it possible for end users to use the same accounts for both OpenStack and the Big Data Software on top of it.

The second part of this talk is about the impact of running virtualized Hadoop on the performance of both OpenStack and Sahara itself. Specifically, we will look at:

  • The progress we have made in improving virtualized I/O performance since the Atlanta summit
  • A comparison of the performance of different Hadoop installations
  • Using Rally for benchmarking
  • Clusters provisioning and workloads execution performance

At the end of this session, you will have a good grasp of what Sahara can do outside of its traditional uses, and a feel for performance capabilities and issues in its use.

Speaker Bios

Sergey Lukjanov

Sergey is the Project Technical Lead of OpenStack Data Processing program ("Sahara", ex. "Savanna"). He has been involved in the project from the first days. One of his main responsibilities is architecture design and community-related work in Sahara. Also he is a top contributor and reviewer of Sahara and he oversees all Launchpad and Gerrit activity. Sergey is experienced in Big Data projects and technologies (Hadoop, HDFS, Cassandra, Twitter Storm, etc.) and enterprise-grade solutions. He implemented HA for Twitter Storm and Sergey is contributing to different open source projects now including Twitter Storm and OpenStack. Also, he's currently the OpenStack Infrastructure core/root team member.

Andrew Lazarev

Andrew Lazarev is a Senior Engineer in the Sahara group at Mirantis. Andrew joined Mirantis 9 years ago and has walked the whole road along with the company, from low level networking for Cisco to Big Data for PayPal and Attensity and finally to open source OpenStack. Andrew started work on OpenStack in early 2013 and now actively contributes to the Sahara project.





Sergey Lukjanov

Principal Software Engineer, PTL of OpenStack Data Processing ("Sahara"), Mirantis

Andrew Lazarev

Senior Engineer at Mirantis

Attend The Summit

Summit Registration

Full Summit Details

(Thank you for all of your input. Voting for presentations at the OpenStack Paris Summit has closed.)