Sorry, you need to enable JavaScript to visit this website.
Home / OpenStack* on Intel® Architecture / Blogs / Kraghuram / 2014 / OpenStack* on Intel® Architecture - Day 2 and 3 at the OpenStack Paris Summit
Blog

Day 2 and 3 at the OpenStack Paris Summit

Author: 
Krish Raghuram

The second and third days of the Summit reinforced the perception that OpenStack is deployable today to provide a tremendous amount of flexibility and agility for the organization, but with a few caveats.

Those who have succeeded, like Numergy, Workday and SAP, have very carefully planned their deployments by taking their own unique situation into account. In the case of Numergy of France (represented by Patrick Debius-Pesquet in the panel session moderated so well by Paul Miller), they had to strike a balance between experience and youthful enthusiasm, going so far as to hire experienced people from other fields "with the capacity to learn and absorb", while at the same time targeting universities for fresh talent. Others like Workday (represented by Carmine Rimi) had to focus on continuous integration and deployment processes to meet their customer demands. 

Still others like SAP have had to carefully consider the existing architecture and internal standards, both at the underlying platform level as well as at the higher level management processes, while sliding a new cloud environment in between. The fact that they've successful deployed back at the Folsom release and kept the environment updated and running at high availability for over 3 years shows that this can be done if the right integration partner (BI Systems in their case) is chosen. They're now able to claim an impressive cost reduction over using a public cloud to deliver the same level of service.

Lastly, one has to be prepared for a time consuming upgrade process, which in turn means they can only happen once every few years, and certainly not every 6 months. The point was reinforced in a Win The Enterprise BOF session run by Carol Barrett of Intel, where end-users and experienced people from the vendor community spoke of the need for "Long Term Support (LTS)" releases from the Foundation. In other words, older releases that would be incrementally improved with perhaps a few new features back-ported from the most recent release. There would thus be multiple "trunks", but this is something the Linux comunity has learned to live with and manage, so there's no reason why OpenStack cannot adopt the same model.

Looking at the traffic at the Intel booth and in breakout sessions of people showing interest in the more advanced scheduling, security, performance and high availability features that we are proposing, it's clear that there's been a big uptick in momentum over the last 6 months. I noticed a large number of meetings taking place in the numerous open meeting areas at the venue (with doubtless many more in the meeting rooms), suggesting we're going to see a lot of activity leading up to the next Summit in Vancouver next May. Over the next few weeks, I and other colleagues in Intel will be blogging about specific features and capabilities that are being driven into the Kilo release for Apr'15

Closing out with kudos to the dedicated teams that put together the demos and presentations for Intel and made this the best Summit for us by far.

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