Austin
April 25-29, 2016

Event Details


Ironic: Managing Thousands of Servers in Production

The challenge of managing physical hardware can be simplified using Ironic bare-metal provisioning, decreasing workload deployment time. In this session, we will address the need for self-service and its benefits over submitting tickets. We’ll also discuss typical data center/infrastructure integration use cases.

We discuss:
- What Ironic is and how it works

- Benefits of using Ironic compared to other provisioning tools

- Challenges and "missing parts" you can encounter using Ironic in production environments, such as:  rack-level awareness for networking (i.e., availability zones for networks); DNS integration; hardware RAID configuration (e.g., extended Ironic Python Agent functionality); server name generation

- Operational issues such as data center networking to accommodate Ironic

Ironic is perceived by many as being immature, since there is a lack of notable experience in using Ironic in production, especially at scale but we’ll share real world large-scale production experience.


What can I expect to learn?

In this session, attendees will learn about:
- What Ironic is and how it works
- Benefits of using Ironic compared to other provisioning tools, such as:
   - Standardized OpenStack REST API
   - Image based provisioning
   - RBAC separations on project/tenant basis
   - VM provisioning
- Challenges and "missing parts" you can encounter using Ironic in production environments, such as:
   - Rack-level awareness for networking (i.e., availability zones for networks)
   - DNS integration
   - Hardware RAID configuration (e.g., extended Ironic Python Agent (IPA) functionality)
   - Server name generation
- Operational issues such as data center networking to accommodate Ironic

Wednesday, April 27, 5:20pm-6:00pm
Difficulty Level: Intermediate
Mirantis, Inc.
Mykyta Gubenko is a Infrastructure Engineer at Mirantis working for the Services department. As an experienced system engineer, he helps Mirantis customers to be successful with Openstack. Mykyta is focused on deployment automation and large-scale openstack projects. FULL PROFILE
Mirantis, Inc.
Alexander Sakhnov is a Senior Software Engineer working for the Mirantis Services department. He has been with Mirantis for more than 6 years, and joined the OpenStack community from the very beginning of the Cactus release. His main activity is helping customers with production OpenStack deployments. FULL PROFILE
Mirantis Inc.