Chameleon is a testbed designed to support cloud computing research. To to support Computer Science experiments ranging from operating systems and virtualization to networking and security, Chameleon provides deep reconfigurability (bare metal, BIOS reconfiguration, console access, etc.). These capabilities are delivered using primarily OpenStack with Ironic: a cloud computing resarch testbed built as a cloud. The testbed is distributed over University of Chicago and TACC and consists of ~400 nodes/10,000 cores and 5PB of total storage to support HPC and BigData experiments, with high-memory, large-disk, low-power, GPU, and co-processor units planned for deployment this year. Chameleon went public in July 2015 and currently supports 600+ users as well as many exciting cloud computing research and education projects. This talk will describe the goals, design strategy, implementation, operation, and the cloud-related research projects taking place on the Chameleon testbed.
This talk describes an exciting, new application of cloud computing: an experimental testbed for Computer Science, targeting BigCompute and BigData research projects; we will describe describe the goals, design strategy, implementation, operation, and the cloud-related research projects taking place on the Chameleon testbed.
In addition, we will describe experiences with larg-ish deployment of Ironic, as well as extensions and adaptations of OpenStack to fit the needs of this application, for example by using the Blazar project to provide advance reservations, or our implementation of snapshotting.
Finally, we will describe the reserach and education projects in cloud computing that users are running on this testbed which, while not OpenStack-specific, may be of broader interest to the Summit attendees.
Registration Opening Soon
Registration will open for you to purchase passes for the OpenStack Summit very soon!
We're excited to see you there!
