Sorry, you need to enable JavaScript to visit this website.
Home / OpenStack* on Intel® Architecture / Blogs / Kraghuram / 2015 / OpenStack* on Intel® Architecture - Day 2 at the Vancouver Summit
Blog

Day 2 at the Vancouver Summit

Author: 
Krish Raghuram

Another day of exciting jam-packed sessions. If the current trend of increased attendance (6000 vs. 4500 in Paris) continues, there's going to be a real challenge for the meeting rooms in Tokyo! Per-session metering and charge-back? Just joking

Mark Collier touched upon the various stages in maturity and adoption, and how OpenStack is set to breakout of experimentation and make the transition to being a "winner", as opposed to becoming a niche offering as it matures but fails to get sufficient levels of adoption. Imad Sousou of Intel pointed out in a matter-of-fact way that Intel is the largest contributor into the Linux kernel, and that track record and expertise allows us to bring in some interesting innovations like Clear Linux, which speeds up deployment of an instance by an order of magnitude. Walk-on sessions by James Penick of Yahoo and Subbu Allamraju of eBay left no doubt that a lot can be achieved if you're willing to invest quality technical skills to the deployment. One got the sense that they were also saying to the community "come on guys, we've ploughed a path, now keep the momentum going with improved capabilities so everyone continues to benefit from this, including us".

While the Paris Summit threw the spotlight on end-user efforts, this one seems to be doing the same on container technology and its place. Talking with some people in the hallways, one could sense a feeling that a container model (orchestrated by Kubernetes or Docker Spawn) could service many emerging cloud applications without the overhead of OpenStack software, and this could slow OpenStack deployment to some extent. On the other hand, we also heard of the benefits of using OpenStack as the control plane across all types of deployments - whether containers, VMs or bare metal.

The Product Management workgroup session run by Sean Roberts of Akanda, Carol Barrett of Intel and Arkady Kanevsky of Dell went into the process and milestones for capturing end-user requirements, getting community review, prioritizing them, and then translating them into blueprints and commitments from developers. They took great pains to emphasize this was not more overhead and work for PTLs but actually a service to them, so their jobs in deciding what to prioritize is made easier. There's also the benefit of being able to show users where each submission is in the process - the projects kicked off within various modules, status of completion etc. - so there could be a more detailed roadmap of capabilities from their perspective and not just from the developers' perspective.

I got to spend some time attending sessions by IBM and VMware, who have not been considered to be the main distributors for OpenStack till now. However, they bring considerable expertise and credibility in supporting large enterprise deployments, and each clearly had their own vision for how cloud computing could evolve, and of OpenStack's place in it. For IBM it was about supporting mutliple hybrid usage scenarios like dev/test, spillover, distributed apps and portability across different geos/regions. For VMware, it was tight integration with existing infrastructures, scalability and strong ecosystem support. Competition is definitely heating up!

Archive

Further Reading

Author: 
Nicole Huesman

This coming week, our team will share how Intel is helping address the requirements demanded by data-centric, compute-intensive workloads quickly growing across data center and edge.

Author: 
Manjeet Bhatia

Devops principles, like continuous integration (CI) and continuous delivery (CD), are attracting considerable attention given their propensity to increase software development efficiency and facili