Interview with Giuseppe Paternò (“Gippa”) , OpenStack board candidate and our collaborator
We had a chat with Giuseppe Paternò (a.k.a. “Gippa”), the author of the introduction to OpenStack we’ve published on the last few issues of GURU advisor.
Giuseppe is Managing Director of GARL (a Swiss security company) and has a long-time experience as consultant and advisor in Europe for cloud-based projects like OpenStack, CloudStack, OpenNebula and Ganeti. He has worked for several companies leader in the sector, like Sun, IBM, Symantec, Red Hat, Canonica and Wind/Infostrada.
Giuseppe, what will be the news in Cloud in 2016, according to your perspective?
Alessandro Perilli, whom we interviewed a few month ago, pointed in particular how Cloud Computing is necessarily a “plural” phenomenon where several different solutions have to come into and collaborate together.
With that in mind, what you feel about OpenStack and the other orchestration solutions (like Cloud Forms) that seem to becoming more and more necessary in order to have a real enterprise solution?
The trend in Cloud is shifting towards applications and multi-cloud. We won’t just have private and public clouds, the trend -yet in 2016, perhaps- is to place our workloads where they better fill or according to the area. And I agree with Alessandro’s opinion because of that.
Cloud Management Platform solutions (CloudForms, ma anche CSA di HP e Scalar) are a very handy tools for multi-cloud environments for sure.
I think that more standardization in resource access, in particular a set of standard APIs, is an added value to achieve a standard platform that can be used by integrators to develop solutions. And I think that the answer to that question is going to be OpenStack.
In my opinion, the “smart” use of applications will be very interesting as well. I dream of am OpenStack dashboard where users will upload their applications and ask to to execute them according to certain workload and service levels. Autoscaling (that is, adapting to the workload), disaster recovery, backup and security will be automatically managed by the platform. We won’t see that in 2016 for sure but projects like Docker, OpenShift and Kubernetes are towards this opinion of mine.
Coming back to our Italy where a few understand the idea of orchestration and Enterprise IT, how far have we been gone in the adoption process of these technologies? Do you see more “geeky” Cloud Providers selling IaaS solutions with a few hosts and a storage, more or less, redundant or do you say that the idea of “cloud architecture” is spreading and platforms like OpenStack (and its competitors) are gaining some momentum?
I’ve been working on OpenStack since 2011 and never like in 2015 I’ve seen so much interest and request for clarifications. However the adoption is very slow and lots are still in the Proof of Concept or internal test phase. If we exclude Great Britain, Italy is in line with the rest of Europe in terms of the general perception of this technology and the adoption of cloud, even if it’s not as quick as I’ve initially thought. It’s true that service provider and outsourcers are the first to understand that, sooner or later, they will adopt OpenStack, but they too have difficulties “digesting” it.
Still, there are some interesting cases in Italy as well: an Italian client of mine, very conservative, who was offered to use the services by Azure as part of the contract by Microsoft, wanted to optimize costs and decrease the requests from the IT department. By using a Cloud Management Portal and automation procedures, now he owns an internal productive zone based on VMware with the “traditional” Oracle and SAP workloads, another internal zone based on OpenStack for testing, development and “easy” Web production, and an external zone on Azure to run public Web sites.
What can you tell us about running for the OpenStack board? Do you think it’s a role where your efforts and your knowledge can really make a difference? What are your potential projects for the future?
I’m running as a candidate part as a challenge, and part because I was pushed by clients and friends working in the OpenStack engineering.
I’d like to point out that I’m coming from the field, even though I have more than 20 years of experience and I’ve worked all around the world. This is really the characteristic that makes me stand out from the other candidates. I want to bring to the board my “on the field” experience, where understanding the needs of the clients and knowing the problems in the actual software is important. I want to be the clients’ spokesman: regional service providers, outsourcers, companies and system integrators.
I’d like to have a foundation more focused on the needs of the market and less inclined towards those vendors that have to sell to fill the needs of the market exchanges and investors.
I’d like the same evolution of Linux in OpenStack, trying to avoid some errors, but, most of all, following its merits and values. Linux, in order to become what it is now, took some years to reach stability in the core of packages. I’d like to see the same evolutive process, if possible in a fewer time, with a stable core, efficient and compact.
I’d also like to see the concept of“stable release”, as it happens with Ubuntu LTS or Fedora vs. RHEL/CentOS. The releases of OpenStack are becoming too much and clients can’t update such a critical infrastructure every 6 months. At the moment switching from a release to a non-consecutive one is really an hassle and system integrators have their issues with integrations. Companies want long-term stability, in particular on main components.