Carrier SDN

Telefónica, Red Hat, Cyan Boost NFV Performance

Telefónica, Red Hat, and Cyan are collaborating on orchestration software designed to maximize performance on networks that use Network Functions Virtualization.

NFV replaces dedicated appliances with virtual machines for network applications such as load balancing and firewalls. The new orchestration software will allow a network operator to deploy virtual machines in specific locations on the network -- a specific datacenter, or rack within a datacenter, or a specific blade on a rack, according to a joint announcement by Telefónica SA (NYSE: TEF), Red Hat Inc. (NYSE: RHT), and Cyan Inc. today.

"Deterministic" is the word the companies use to describe the ability to place virtualized network functions precisely on the network, as opposed to just tossing the VNFs into the cloud and letting the cloud stack determine where the VNFs should run.

"Deterministic placement is about leveraging technology to do very specific placements so you can extract the optimum performance leveraging the optimum infrastructure," says Nirav Modi, director of software innovation at Cyan.

Red Hat plans to incorporate the orchestration software into the open source OpenStack cloud stack, as well as including it in Red Hat's own software and services, says Radhesh Balakrishnan, general manager, virtualization and OpenStack at Red Hat.

Cyan plans to incorporate the software into its own Blue Planet orchestrator. Cyan is still trying to figure out whether and how to open source the orchestration component; the industry doesn't have an open source NFV implementaiton to support the software, says Modi.

Telefonica is driving the development effort, in conjunction with Cyan and Red Hat, says Modi. I talked with Cyan and Red Hat, but I couldn't get a hold of Telefonica. I'm working on that.

— Mitch Wagner, Circle me on Google+ Follow me on TwitterVisit my LinkedIn profileFollow me on Facebook, West Coast Bureau Chief, Light Reading. Got a tip about SDN or NFV? Send it to [email protected]

Diego Lopez, Telefonica senior technology expert, is one of the featured keynotes at Light Reading's Big Telecom Event (BTE), which will take place on June 17 and 18 at the Sheraton Chicago Hotel and Towers. The event combines the educational power of interactive conference sessions devised and hosted by Heavy Reading's experienced industry analysts with multi-vendor interoperability and proof-of-concept networking and application showcases. For more on the event, the topics, and the stellar service provider speaker lineup, see Telecommunication Luminaries to Discuss the Hottest Industry Trends at Light Reading's Big Telecom Event in June.

More about NFV:

njmodi 6/2/2014 | 8:50:41 PM
Re: How does this work? Hello,

I would be happy to explain this engagement in more detail if you are interested. Please contact me via email: [email protected]

Atlantis-dude 6/2/2014 | 8:01:25 PM
Re: How does this work? If someone looks at the solution that will appear in open source (OpenStack) how will they know the requirement that it is for ?
njmodi 6/2/2014 | 2:39:44 PM
Re: How does this work? The requirements are internal to Telefonica and not going to be published.
Atlantis-dude 6/2/2014 | 2:31:48 PM
Re: How does this work? Where is Telefonica publishing the requirements ?
njmodi 6/2/2014 | 12:28:31 PM
Re: How does this work? In this particular engagement, Telefonica is driving the requirements based on their own technical findings, and Cyan and Red Hat are collaborating to provide a solution. The software enhancements are going to be submitted back into the open-source (OpenStack) community.
Atlantis-dude 5/31/2014 | 2:07:21 AM
How does this work? So a service provider is driving software development and system vendors are pulling that software into their offerings ? 
njmodi 5/30/2014 | 9:03:29 PM
Re: Placement advantage Hello Carol,

Let me try illustrate this concept in action. Consider a virtualized network function like an Evolved Packet Core (EPC). This type of VNF typically has a composite (and clustered) architecture. By this I mean that the VNF, when deployed, will very likely be comprised of multiple VMs (VNF components) that will be interconnected and together make up the EPC function. In the case of an EPC, VNF components that represent active/standby controller functions (or database, etc.) must be placed on physically separate infrastructure to ensure the high-availability (HA) that network operators require for these mission-critical applications. This would be an example of specfic placement - and would fundamentally ensure that HA is achieved. If the VNF components are placed non-deterministically, you could have a situation where the active/stanby controllers are actually within the same blade, and in that case you don't really have equipment-level redundancy.


Carol Wilson 5/28/2014 | 3:27:45 PM
Placement advantage Mitch, what's the advantage of specifically placing a virtualized function?
Sign In