#SDBigData Meetup #21

5:45 - 6:45
Socialize over food and adult beverages

6:45 - 730: Learn how to get a functioning Hadoop cluster on bare metal
Greg Bruno, StackIQ (https://www.stackiq.com/)

7:30-8:15:

Containerizing Kinetica Real-Time Analytics with NVIDIA – Docker – Made Easy!
Jacci Cenci, Nvidia (http://www.nvidia.com)

-------------------------------------------------------------------------------------------------------

Speaker Details:

Speaker: Greg Bruno (https://www.linkedin.com/in/greg-bruno-46558584/), VP Engineering and Co-founder, StackIQ

Description:

"Step 1 of every Hadoop vendor’s documentation reads something like this: “First install a cluster.” Without a consistent group of installed machines, a Hadoop installation is prone to failure. Architected, developed, and built completely in the open, the Hortonworks Data Platform (HDP) provides Hadoop designed to meet the needs of enterprise data processing. The deployment of HDP on a cluster is a non-trivial task. And while Ambari is used to deploy HDP on a cluster, Ambari itself needs to be set up on a cluster too. Stacki automates the deployment of Ambari in a few simple steps.

Stacki is an open source bare metal provisioning tool that installs machines to a ping and a prompt enabling the consistency and configuration required for modern applications, including Hadoop.

The StackIQ (http://t.sidekickopen04.com/e1t/c/5/f18dQhb0S7lC8dDMPbW2n0x6l2B9nMJW7t5XX48q5T4zW1q0Qn21qwvvvVQBb2Y56dS7ldcq1r202?t=http%3A%2F%2Fwww.stackiq.com%2F&si=4747341534330880&pi=3f9d2566-b39a-48c2-ee54-b3c1b8113a1b) engineering team recently released an open source Stacki Pallet for Hortonworks (http://t.sidekickopen04.com/e1t/c/5/f18dQhb0S7lC8dDMPbW2n0x6l2B9nMJW7t5XX48q5T4zW1q0Qn21qwvvvVQBb2Y56dS7ldcq1r202?t=http%3A%2F%2Fstackiq.com%2Fhortonworks%2F&si=4747341534330880&pi=3f9d2566-b39a-48c2-ee54-b3c1b8113a1b), which provides the software necessary to easily deploy Ambari and then HDP on a cluster. This presentation will demonstrate how to the Stacki Pallet for Hortonworks can be used to give you a functioning Hadoop cluster on bare metal. You will learn how to set up Stacki (http://t.sidekickopen04.com/e1t/c/5/f18dQhb0S7lC8dDMPbW2n0x6l2B9nMJW7t5XX48q5T4zW1q0Qn21qwvvvVQBb2Y56dS7ldcq1r202?t=http%3A%2F%2Fstackiq.com%2Fdownloads%2F&si=4747341534330880&pi=3f9d2566-b39a-48c2-ee54-b3c1b8113a1b), the Pallet, Ambari, and then install Hadoop on a running cluster. You can download the necessary ISOs for the Pallet and view the documentation on the GitHub Repo: https://github.com/StackIQ/stacki-hdp-bridge "

Speaker: Jacci Cenci, Nvidia

Join us to learn how Nvidia and Kinetica are revolutionizing GPU computing and working together to make application deployment easier for your enterprise.
Docker containers are often used to seamlessly deploy CPU-based applications on multiple machines. With this use case Docker containers are both
hardware-agnostic and platform-agnostic. This is obviously not the case when using NVIDIA GPUs when deploying applications like Kinetica in a container.
Kinetica is using specialized GPU and NVLINK hardware and software. Enterprise software applications require the installation of the NVIDIA drivers. To make the Docker images portable while leveraging NVIDIA GPUs, see how nvidia-docker wraps the Kinetica application images in a way that makes it agnostic of the NVIDIA driver in our demo.