“To boldly go where no one has gone before!”
Those words still echo in my mind as I remember watching the old Star Trek shows from yesteryear. It rings of adventure, of exploration, and of never settling for the known state of things.
It is these words that come to mind when I think of the new Voyager technology that is coming to market, which is designed to go boldly where no other technology in the Packet and Optical world has gone before. Voyager is the industry’s first combined routing, switching and optical platform all combined in a 1 RU footprint. The unique combo sets out to unifying both IP and optical to massively reduce complexity and costs. It will boldly transform the data center interconnect of today.
But it’s also the first open offering in the optical space. Cumulus is bringing its networking with S.O.U.L. (Simple. Open. Untethered. Linux) moxie and applying it to the transport and data center interconnect markets. This disaggregated solution dramatically reduces the cost of the current proprietary stack. It’s a solution with multiple players…
- Facebook – Donated the design to TIP
- Cumulus – The open network operating system (NOS)
- Celestica – The hardware provider
- Broadcom – The ASIC chip provider
- Acacia – The optics provider
- ADVA – The management/solution vendor
- Zeeta Networks – Orchestration
And that has piqued the interest of companies globally, who seek a fundamentally better way to manage switching and DWDM transport. For example:
“We’re interested in finding ways to drive down the cost per bit across our backbone. In working with our colleagues at the New York State Education and Research Network (NYSERNet), we see promise in disaggregated solutions like Voyager to help us get there in a flexible and scalable way.” — John Moore, AVP for Research and Infrastructure Strategy, Internet2
New York State Education and Research Network:
“We are excited to have so much bandwidth potential in such a compact form factor, coupled with the capabilities of Cumulus Linux. We anticipate significant growth in research traffic, and our partnerships with Cumulus and ADVA, along with our participation in TIP, will help us unlock the potential of our network to handle those demands.” — Bill Owens, CTO of NYSERNet
GRNET, the Greek research and technology network, has successfully completed a joint field trial involving Voyager.
Knowing the insatiable demand for internet, bandwidth-intensive apps, and the rise of edge computing, companies like Vodafone are seeking for a scalable, yet cost-effective long-haul interconnect solution. In their announcement, Vodafone mentions that the goal of their live trial of Voyager in Spain was to “showcase the future of applying a disaggregated model to optical networks,” as well as providing more flexibility to handle the dynamics of real time modulations.
This trial at Vodafone shows how Voyager can be implemented over an existing optical infrastructure. The results of the trial were impressive:
- Able to deliver 800 Gigabits per second (GBPS) per rack
- Ability to dynamically adapt the system modulation as fibre conditions changed
- Using SDN principles to conduct real-time monitoring of the optical line performance
- Proved that “our live network can set up optical services and keep them running”
- Reduced unnecessary and lengthy customer service interruptions
- Improved utilization of their network
These are impressive results for Voyager. But there is a great backstory behind how Cumulus brought all this together for Vodafone. What Cumulus enables is a full Layer 1 through Layer 3 implementation to help set the various optical level specs (think modulation formats, transmission power, FEC configuration, wavelengths, etc.), as well as to configure port bundles, VxLANs, routing, ACLs, etc., for the L2-L3 layers.
Our co-founder and CTO, JR Rivers, was a big part of that backstory, so I asked him about it. Here’s a deeper dive from JR that gets into the process of getting Voyager ready to test at Vodafone:
“The Vodafone-Voyager project highlighted how a bunch of simple elements that leverage each other can create an incredibly powerful result. The team consisted of people from Cumulus Networks, ADVA Optical, Zetta Networks, Facebook and Vodafone. Facebook provided the hardware, Cumulus the NOS and element API, and Zetta the SDN controller. This all came together over the course of a couple months while each of the components were being separately developed. On the Cumulus side, we developed the REST API and data model, along with a simulator so that the Controller team could do their development in parallel with the platform team. Initial integration occurred in Bristol, UK where the all of the components (along with the test equipment) came together for the first time. As you’d imagine there were a few learning curves… many of which were tied to this marriage of IP and optical.
For instance, some team members with a long history in the L1 optical world spent about a day convinced that there was an artificial loopback in the system…only to find out that when you put an interface into a bridge, you get Spanning Tree Protocol and Link Layer Discovery Protocol frame that come out at a regular cadence. And then there was the part where we needed to put meaningful Ethernet MAC addresses on test packets from optical BER testers. It was really cool watching people learn each other’s disciplines and breaking down traditional silos.
We also were able to see how Cumulus Linux helped the development/integration process in unexpected ways by opening up credentialed ssh/vpn tunnels into lab equipment to allow development to occur remotely. The second integration occurred in Madrid with a fresh set of Voyager platforms and it went incredibly smoothly! Even though all functionality was not quite complete, everything that was supposed to work performed as expected; we just needed a bit more polish. A couple weeks later, the final trial was done in Vigo on an test optical plant complete with noise generators. The team allowed three days for the testing, so flights and hotels were booked and schedules adjusted accordingly. Imagine how bummed the team was when all testing was completed in the first day (check out Vigo if you don’t get the joke).”
I’m sure they all had a wonderful time in Vigo! But as you can see, things worked as smoothly as it did because of the unifying power of Linux! Vodafone could utilize the Zeeta orchestration controller to manager Voyager because of the open nature of Cumulus Linux running on top of Voyager. Simulated environments were created because of Linux, and an API written for the connection to happen, because of Linux.
In other words, none of these results would have been possible without Cumulus in the middle of everything.
To learn more about this union of packet and optical, and how it can help you, take a deeper look into all the benefits of Voyager. Or, sit back, relax and enjoy our how-to Voyager video series! You can also check out the official press release here.