Configuring, deploying, and managing a fully operating cloud environment is not for the faint of heart. Despite the popularity of Openstack, the premier open source cloud environment, “it's still widely known for being hard to install and deploy.But, installations and deployments don't often fail as a result of OpenStack itself; more often, it's due to the amount of time and resources it takes to get it up and running. And, aside from tech-savvy early adopters, most operators with minimal cloud configuration experience find deploying, scaling and operating OpenStack overwhelming.”
OpenStack has become an integral component in helping operators realize their NFV plans. By the numbers, the OpenStack Foundation claims:
- ~30,000 Individual Members from over 170 countries around the world
- Over 600 corporate sponsors, donors and supporting organizations
- ~20 sub-projects
Industry participation at their events has also grown – by over 2 orders of magnitude from 2010 to 2016. So while there are issues voiced about its ability to perform in carrier grade deployments, there’s a lot of momentum behind its further adoption.
OPNFV (Open Platform for Network Function Virtualization) is basically an Openstack deployment framework, with emphasis on the networking side. It describes itself as “a carrier-grade, integrated, open source platform to accelerate the introduction of new NFV products and services”. It’s also getting a lot of notice by service providers and NEMs alike. There are currently 55 member companies involved in OPNFV with over 200 active open source developers working on 47 open source projects.
Even with the OPNFV umbrella covering all of the projects involved in an NFV deployment, there are many, many possibilities for configuring a working system. How do you find your way through the maze? There is no single answer, but here are some thoughts from one engineer’s wanderings. The blogs that follow are a brain dump of the various things encountered and learned in my OPNFV adventures over the last 9 months. This has encompassed two OPNFV releases – the original “Arno” and the current “Brahmaputra” release. In addition, I have worked through deployments on two different hardware environments – a set of 1U rack mount servers and an HPE BladeSystem with Gen9 blades.
The end results have been two functioning deployments that support Dialogic’s Powermedia™ XMS Media Server and Media Resource Broker (MRB). They will be used for developing the virtual infrastructure that surround our NFV products, simulating similar Telco customer environments, and as a flexible basis for our company-wide QA and test efforts.
Since almost all of what I do on a daily basis is expected to actually function, this series of blogs will focus on the practical rather than the theoretical. We have no lack of NFV marketing prose extolling the virtues of the virtualized network. This will not add to that. What I hope to give you is some hard information – at as high a level as possible – on what worked (and didn’t work) for me, in the hope that it will assist you and your company in going down the same virtual road.
One might ask – if a full cloud is so difficult to achieve, why not just use a hypervisor like KVM, Xen or VMWare, surround it with a light duty management GUI like virt-manager, script some of the needed network-level virtualization functions and be done with it? Why is a full VIM (Virtualized Infrastructure Manager) needed? Right now, using ETSI specs alone as guidance for deploying a Telco Cloud is not enough. Simply put – you do not want to even think about embarking on this sort of project on your own.
You need help to manage the complexity and to not re-invent the wheel in doing so. There’s a large community, as I pointed out earlier, behind OpenStack and OPNFV, and a huge amount of work has already gone into the latter and its underlying projects. So let’s all take advantage of it. Let us know about your involvement in these projects and share with us your insight in moving to the cloud by tweeting us @Dialogic. Stay tuned as we further demystify OPNFV with follow-up blogs on the following:
- Intro to Building a Carrier Grade Cloud Operating System Using OPNFV (OPNFV Demystified Part 1)
- The Components of OPNFV (OPNFV Demystified Part 2)
- Getting Started with OPNFV - How Should I Prepare? (OPNFV Demystified Part 3)
- Real-World OPNFV Implementation – The Hardware (OPNFV Demystified Part 4)
- Real-World OPNFV Implementation – Traps and Tips (OPNFV Demystified Part 5)
- Real-World OPNFV Implementation – Networking (OPNFV Demystified Part 6)
- Why Automation Is Key to Your OPNFV Deployment (OPNFV Demystified Part 7)