OpenStack Summit Tokyo – Day 3

Hello again from Tokyo, Japan where the third and final day of OpenStack Summit has come to a close. As with the previous days of the event, there was plenty of news, interesting sessions, great discussions on the show floor, and more. All would likely agree that the 11th OpenStack Summit was a rousing overall success!

Like day 1 and day 2 of the event, Red Hat led or co-presented in several sessions.  Starting us off today, Erwan Gallen, Red Hat’s OpenStack Technical Architect, participated in a panel and helped provide an Ambassador community report. Among other things, the group of OpenStack ambassadors introduced several improvements over the past six months, since the last OpenStack community release (Kilo), and shared many of their overall feelings and experiences about the community.

Mark McLoughlin, Red Hat’s OpenStack Technical Director, then gave an interesting talk entitled The Life and Times of an OpenStack Virtual Machine. Delving deeper than a simple, abstract narrative of initiating a Launch Instance in OpenStack’s dashboard, Mark detailed the technologies involved behind the scenes to allow for this. By the end of the session he had fully explained how OpenStack provides a running VM that a user can access via SSH.

Erich Morrisse and Massimo Ferrari then teamed up to give a presentation entitled Elephant in the Room: What’s the TCO for an OpenStack cloud? The pair discussed recommended tools and methods to provide the best quantifiable answer to this challenging question.They introduced a TCO model, and a method of using standard financial accounting practices to build a basis for calculating the value that an OpenStack cloud provides.

In his talk, Debugging the Virtualization layer (libvirt and QEMU) in OpenStack, Kashyap Chamarthy, a Senior Software Engineer, explored virtualization drivers (e.g. libvirt, QEMU/KVM) and explained how they are the core part of the OpenStack Compute layer. He then acknowledged that an OpenStack environment is challenging enough to debug as it is, but even more so when multiple Compute nodes (and thereby multiple libvirt daemons and QEMU instances) are involved. In the end, he explained how libvirt and QEMU provide a rich set of debugging controls that allow users to query or modify the state of virtual machines in distress.

Josh Durgin, Senior Software Engineer, and Sèbastian Han, Senior Architect, then delivered a presentation to a packed room entitled Ceph and OpenStack: Current Integration and Roadmap. In their talk, they described the current state of the integration of Ceph into OpenStack and where it is heading in the future. In particular, they explained that with the Juno release, we reached a critical step in terms of overall features, robustness and stability. The Kilo release, they confirmed, is even better, and Liberty is even more promising–but the best is yet to come.

Steve Gordon, Senior Technical Product Manager, never fails to impress with his presentation titles. In his session Dude, this isn’t where I parked my instance!?, Steve explained that OpenStack Compute provides a number of facilities for moving instances around, but affirmed that it’s not always obvious how they differ from each other. He then detailed the differences between each of the available options including evacuations, cold migrations, and live migrations as well as the internal mechanics of each, including some of the ways they can differ when using different hypervisor backends. He also discussed the prerequisites for enabling each method and the optimal configurations for providing the right combination of security and performance.

Later in the day, Mike McCune, Red Hat Senior Software Engineer, joined Andrey Brito, Professor from the Universidade Federal de Campina Grande (UFCG) and Telles Nobrega, a Software Engineer and Masters Student at the Laboratório de Sistemas Distribuídos, to discuss Sahara+Storm: real-time data analytics in OpenStack. In their talk, they detailed the Storm plug-in for Sahara and guided attendees through the essential steps to set up a scalable, real-time data processing application. They also shared plans for improving real-time data processing in OpenStack.

And finally, last but certainly not least, Sean Cohen, Principal Product Manager from Red Hat, teamed with Akshai Parthasarathy from NetApp, and Thomas Becktold from SUSE, to give a talk entitled Manila – An Update from Liberty. The three provided a demonstration of how to install Manila using popular distributions, and discussed proposed blueprints for the Mitaka release.

It goes without saying that it was a busy final day at OpenStack Summit. As mentioned previously, the final day, and the event overall, was jam-packed with a wide variety of best practices, industry uses cases, community news and updates, and much more.  Like all OpenStack Summits, it was an extremely informative event, and also lots of fun!  If you missed our previous daily recaps, we encourage you to read our blog posts from Day 1 and Day 2.  And for those who were present, we hope you enjoyed the event and found time to visit the Red Hat booth, as well as network with friends and colleagues from around the world. We’re already counting down the days until the next OpenStack Summit — we hope we see many of you again in late April 2016 at the OpenStack Summit in Austin, Texas!

For continued Red Hat and industry cloud news, we invite you to follow us on Twitter at @RedHatCloud or @RedHatNews.

Signing out from Tokyo.  Sayōnara (さようなら)!