Fog and edge computing score big on the tech jargon bingo boards of today.
SuperuserTV talks to Adrien Lebre, who co-chairs the OpenStack Fog, Edge and Distributed Computing Working Group, about why it’s important and how you can get involved. Lebre is a researcher at Inria, France’s national institute for computer science research and applied mathematics. Below is an edited partial transcript of the interview.
What is fog, edge and massively distributed computing? Why is it important right now?
During the last couple of years, there’s been a trend towards building mega data centers by companies like Google, Microsoft, and Amazon. The idea is that you build a mega data center with thousands of servers to cope with the demands of cloud computing. Unfortunately, with the new usage trends — internet of things, tactical internet — these mega data centers cannot satisfy the latency needs for these applications. So,we need to propose a new model that will be able to satisfy all this latency-critical requirements. The idea is to deploy smaller data centers, but closer to the end users, leveraging the internet backbone. Each network point deploys a couple of servers and those servers can satisfy the needs in terms of competition, storage, etc. for these new kinds of applications.
Can you give us a couple of examples of how it would be used, in layman’s terms?
A typical application: you have your smartphone, and it has some limitations in terms of capabilities. Let’s say you want to watch a TV show in 3D on it, the idea is to use the competition that we provide with this edge infrastructure, so instead of running the app on your smartphone, it runs on the closest edge cloud.
A second example is the medical field — think tele-surgery applications. Say you have a doctor who needs to perform surgery remotely. In this case the latency is critical, so you have to provide cloud computing capabilities much closer to these end users.
What role does OpenStack play?
When we started with this initiative, it was from scratch. We took a white paper and started thinking about how a massively distributed architecture that’s deployed close to the end user would look like…
When we started thinking about developing a proof-of-concept, we discovered it’d require a lot of engineering work: enter OpenStack. Instead of re-inventing the wheel we chose to join the community to leverage the rich ecosystem and see how to build on it for our needs.
We created a dedicated working group — the Fog, Edge and Distributed Computing Group — in OpenStack to gather and federate all the developers working on these issues and with the goal of creating this new operating system.
Catch the entire interview with Lebre below.
Here are some ways you can shape what’s next in fog and edge computing:
- Sign up to the openstack-dev mailing list and look for posts with “[Massively distributed]” in the subject line.
- Take part in the bi-monthly meetings on irc #openstack-distributed suggest your agenda items and take part in current discussions (Time should be defined according to the different timezone of participants).
- Share particular use-cases or superuser stories
Review specs and provide your input
- Email Adrien Lebre <adrien.lebreATinria.fr> with your suggestions, questions, …
- Check out past meetings: https://etherpad.openstack.org/p/massively_distributed_ircmeetings_2016