|By Roger Strukhoff||
|April 28, 2014 12:11 PM EDT||
Sandro Mazziotta (pictured below), Senior Director of Product Management at Paris-based eNovance, attended the recent Red Hat Summit in San Francisco and answered a few of our questions about what's going on with the company and with OpenStack.
Here's the second part of our interview with him. The first part is also available at Cloud Computing Journal.
Roger: You mentioned there' a fast-moving and growing market opportunity for mobile on OpenStack, challenged by getting the right data in real-time to mobile devices. What specific opportunities have you seen, and can you provide an example of a project with the challenge of accessing the right data in real time?
Sandro: The growth in mobile on OpenStack is being fueled by several major industry trends such as the emergence of smartphones and smart devices like tablets, the improved quality and performance of radio networks, the emergence of new mobile use cases like gaming, Internet of Things and social media and lastly, the maturity of cloud computing technology as a suitable platform for business critical applications.
On this last point, in particular, what we've seen recently in the marketplace is OpenStack differentiating itself as the platform of choice for network operators to deploy Network Function Virtualization (NFV) infrastructure.
In fact, 9 out of 10 operators have selected OpenStack as the target technology for deploying NFV. The ratio is so high because open source is key for their overall strategy and ability to retain control of their networks. No operators would leave the destiny of their private cloud infrastructure it if was not based on open source software.
Roger: And your company?
Sandro: At eNovance, we're involved in several projects helping to launch large scale OpenStack deployments that enable various NFV use cases. All of these cases are quite complex and always require fine tuning of the OpenStack code on a continuous basis.
As an example, when a customer requires a particular feature that is not yet available in OpenStack, we contribute it first upstream so that it will be supported by all distribution vendors in the coming release. This is the way OpenStack was envisioned to be continuously integrated, continuously deployed.
Roger: You mention that OpenStack is the only open-source cloud management stack to deploy Big Data apps in the cloud. What makes it so? What are its advantages compared to other open-source platforms, and compared to some of the proprietary ways of going about this?
Sandro: Big Data is a unique use case because it's so costly to move data, and you have to decide early to deploy in a public or private environment. This is a difficult decision, notably for enterprises when they build new applications or transform legacy applications.
OpenStack is the only open source platform that gives you the freedom to 1) either rely on public service provider or 2) to deploy on premise in a private cloud. Mobile and Big Data applications are by design better suited to be developed in a cloud architecture to benefit notably from elasticity, and distribution features offered by cloud model.
Roger: Why do you think this is true?
Sandro: For the purposes of efficiency, these applications need to store and retrieve data where the application is deployed. Until now, developers of these mobile applications were lacking easy to use data services in OpenStack. However, with the two latest release. Havana and Icehouse and most notably with the introduction of Trove (SQL and NoSQL Database as a Service) and Sahara, formally known as Savanna, a Hadoop implementation in OpenStack, the OpenStack community is quickly starting to address these challenges.
While these new data services are not specific to OpenStack and are available for all cloud computing platforms (AWS, Google, Azure, etc.), OpenStack is the only open source alternative on the market today, and we believe the most viable framework to meet their current and future needs.