Chapter 7

Why networking with S.O.U.L is different

Why networking with S.O.U.L is different

The benefits of networking with S.O.U.L

By bringing SOUL to your data center, you'll unlock countless operational and financial benefits for your data center. Benefits include:

  1. Cost Reduction – Drastic reductions in costs by disaggregating software from hardware, and the ability to negotiate the best price for bare metal switches.
  2. Choice – Build your network to match the needs of your business, and use the network to drive innovation — whether it is around the routing protocol, with the automation tools or about being able to rely on Linux as the common language of the data center.
  3. Agility at scale – Implement DevOps practices and make your operator-to-switch ratio the same as your operator-to-server ratio.
  4. End-to-End Automation – Achieve end-to-end workflow automation, not just automated configuration.
  5. Host-to-Switch Insight – With deep fabric insight, teams can obtain complete visibility across the fabric for enhanced testing, change validation and troubleshooting

How others compare

Another way to look at this is to compare other networking vendors to just a few of the S.O.U.L. elements we have spoken about. No matter how you slice it, others just don’t compare.

The SOULless Competition

S.O.U.L. in the enterprise

By making the network simple, open, untethered and Linux-based, organizations set themselves up for all the benefits of web-scale networking. It becomes the baseline to build a modern networking practice for the digital age.

S.O.U.L. is the foundation of the modern, web-scale network

Below are a couple of examples of companies who are getting amazing results by implementing the various tenants of S.O.U.L.


As a small organization focused on 3D-CAD drawings, Shapeways only has a handful of racks of equipment, but its team’s focus is DevOps and using Linux as a backbone. They are all Linux administrators and like Cumulus because of our Linux focus. They appreciate the fact that they can operate their switches just like their servers, and that their operator to switch count doesn’t change.

Shapeways doesn’t use AWS because of cost inefficiencies due to the amount of storage and compute they required. As their network has grown, they haven’t had to hire any additional network engineers. And with an untethered solution, they have choice and freedom to run their data center the way that works best for them. Instead of using VXLAN, they chose to use routing on the host. Instead of using Ansible, they went with Puppet. They’ve taken these DevOps tools and this monitoring capability to heart.

Athena Health

Athena Health provides network-enabled services for healthcare and point-of-care mobile apps to drive clinical and financial results for its hospital and ambulatory clients in the United States. Athena was looking for a way to scale their network in a more efficient, easier way. When they land a new contract, they often need to build out three, six or even eight racks of equipment with very little notice. Connecting all this equipment became a massive, manual effort to get the network up and going. They needed a better way to scale quickly to meet this demand.

Cumulus provided an answer. Because Cumulus is Linux based, they can automate the process of provisioning the device, so it is off and ready to go in a few minutes. Thus, simplifying the scaling process. Additionally, they loved the open and disaggregated model from Cumulus that allows them to negotiate over the price of licensing of the HW vendor. And when the hardware shows up, they can just apply the software licenses to the devices to get them operating quickly.

Since choosing Cumulus, Athena Health has been able to reduce time to rack a switch from 45 minutes to 3 minutes, and improve overall operational efficiency from 18 hours to a mere 30 minutes. And on top of all that, they use Cumulus VX for testing and training so they can push other changes through the simulated environment and validate what is going to happen and what is expected behavior.


NTT’s Software Innovation Center (SIC) needed a flexible, open network that allowed rapid innovation while controlling their TCO. Yet layer 2 complexities like network loops, rogue DHCP servers, and mis-configuration of IP addresses held them back. They sought to move to L3 technologies to support applications that used custom protocols like GoBGP, while supporting L2 applications with EVPN/VXLAN. “The openness which Cumulus Networks brings to the network world has huge impact to us,” states NTT SIC. With Ubuntu on the compute nodes and Cumulus Linux on the leaf and spine switches, NTT SIC was able to leverage their existing Ansible knowledge and automation experience to automate their network similar to their compute environment.

Next chapter