There a number of ways that Cumulus makes managing the network easier. In fact, this is really our sweet spot. We break them down into the 4 E’s of Simplicity — Easy to Start, Easy to Use, Easy to Scale and Easy to Try.
Here are two examples of how we help you get started with Cumulus right from the very beginning.
Zero Touch Provisioning (ZTP) – ZTP enables network devices to be quickly deployed in large-scale environments. On first boot, Cumulus Linux will invoke ZTP, which executes the provisioning automation used to deploy the device for its intended role in the network. The provisioning framework allows for a one-time, user-provided script to be executed. You can develop this script using a variety of automation tools and scripting languages, providing ample flexibility for you to design the provisioning scheme to meet your needs. Sound familiar? If you've used ZTP before, you know how easy it makes configuration. But since Cumulus Networks runs on Linux, you can do so much more with this automation script. You can use it to add the switch to a configuration management (CM) platform such as Ansible, Puppet, Chef, CFEngine, SaltStack or possibly a custom, proprietary tool. With ZTP, you can reduce provisioning time from weeks or months to now seconds and minutes. Additionally, you can easily swap out entire switches (say, if one fails). The whole switch can be replaced and provisioned easily because the configuration info lives in Git instead of on the box, making for easier revision control and enhancing collaborative configuration. This comes at no additional cost with Cumulus Linux.
Dynamic cabling and validation – Data center cabling is a huge pain and a time-consuming effort. Prescriptive Topology Manager (PTM) is a cabling validation tool that comes free as part of Cumulus Linux. PTM simplifies this process by validating that all the cables are in the right place. PTM leverages standard Linux and open source tools and protocols for ease of use, and interoperates with any network devices or server hosts that support industry-standard LLDP. Other vendors will have a GUI that shows you how to cable, but it won’t validate if it was done correctly.
The ease of managing Cumulus solutions makes it a cinch to adopt. Below are some key examples of what we mean.
Our Network Command Line Utility (NCLU) makes Cumulus Linux easily usable and accessible. It was designed to quicken the learning curve so all network engineers can use it, while integrating with and still supporting the traditional Linux methods. This command line interface for Cumulus Networks products simplifies the networking configuration process for all users. It provides consistent access to networking commands directly via bash, thereby making configuration and troubleshooting simple and easy — no need to edit files or enter modes and sub-modes (of course, unless you want to).
NCLU does a lot more than traditional command line interfaces by:
Exterior Border Gateway Protocol (eBGP) is without a doubt the most scalable routing protocol, which made it a popular choice for large-scale deployments from service provider networks to data centers. Its only real challenge is the problematic configuration process. We’ve transformed the process with eBGP unnumbered, a new innovation to the networking landscape. eBGP unnumbered provides organizations a routed fabric through a leaf and spine network by running BGP through the leaf and spine. With eBGP unnumbered, you don’t have to assign IP addresses to any of these links. In other words, it eliminates the nightmare of IP address management with all those /30s or /31s. It basically allows IPv4 addresses to be advertised over IPv6 sessions. In an average data center, that can save a /24 worth of IP addresses that would otherwise be wasted just on infrastructure links. You simply plug the following into NCLU — “Define the interface” — and the peering automatically comes up and exchanges routes. Super simple. Super easy.
As major contributors to this open source project, FRR is Cumulus’ solution to do routing on the host. FRR provides the ability to speak eBGP down to a server. It is open source software designed by Cumulus Networks and other contributors; it was created to streamline the routing protocol stack and to make engineers’ lives that much easier. It provides Layer 3 connectivity throughout a data center, from the spine switches and leaf switches all the way down to hosts, virtual machines and containers. It is a perfect solution for things like distributed storage, or microservices and containers if you’re using things like Docker, Kubernetes, OpenShift or Mesosphere. Routing on the host provides a highly scalable, unique solution for containers and OpenStack environments.
The visibility achieved through Cumulus NetQ is extremely powerful. In any network, failures happen all too often. When they do, you usually ask: 1) Where is it coming from? 2) How widespread it is, or what is its blast radius? and 3) When and what was the last known good state? With any other vendor, it would take a box-by-box approach to figure out the issue. You might have to go through 20-50 switches to find it. The more time you go through trying to figure out the problem, the more dollars are lost. Enter Cumulus NetQ.
NetQ gives you a very quick way to find out what the problem is, help pinpoint the actual problem and find the last good network state. We call this “actionable insight.” Your network doesn’t stop at your ToR Switch — it actually starts at the host level, meaning your troubleshooting boundary doesn’t stop at the switch. That’s why NetQ runs an agent on the Linux host and on the switch. Having the agents on both provides a much broader, comprehensive view.
NetQ gives you all this insight right from your laptop! When you run a network command from your laptop or your switch, you can query the entire fabric from wherever you are. Plus, NetQ integrates with Slack. If you get a 2:00 AM pager call, you receive all the communication of the switch in the Slack channel to help ID the problem. The key is that you can get notified and then troubleshoot anytime, anywhere.
Gone are the days of having to buy a SDN controller that is proprietary, complex or a scalability bottleneck. Unlike the SDN controller-based options in the marketplace today, Cumulus Networks allows you to completely customize your network based on your needs and your budget. You can leverage existing automation tools, existing talent and existing processes to fully automate a flexible web-scale network.
Cumulus Networks sees EVPN as a universal software-driven gateway to make networking management more simple and agile. As a next-generation end-to-end control-plane solution for VXLAN tunnels that uses the BGP routing protocol, EVPN provides high scale, redundancy, traffic engineering, multi-tenant separation and fast convergence for host and VM mobility — all while interoperating between vendors by featuring both asymmetric and symmetric routing methods. You can learn more about it here.
Cumulus created another open tool called Virtual Routing & Forwarding (). VRFs will isolate routing tables between tenants, allowing for multi-tenancy within the environment. Now you can take a device like a firewall and have the firewall or any policy device in my network provide enforcement between the different tenants. Cumulus EVPN uses the VRF capability integrated with BGP technology to separate tenants within a data center. Each tenant can be mapped to a unique VRF, providing layer 3 isolation of their network traffic.
Cumulus is the only solution that allows you to affordably build and efficiently operate your network like the world’s largest data center operators — whether you have six racks or six data centers. Cumulus provides solutions for both scenarios, with fixed form factor switches as well as chassis based leaf/spine networks.
We recommend using the Clos or leaf-spine topology for predictable scale-out bandwidth. The market is moving to these L3 architectures because you can predictably scale out bandwidth. The leaf-spine topology improves networks — making them more scalable, more reliable and more predictable. If more bandwidth is required, simply add more spines. If more servers are required, add more leaf switches. In this architecture, devices now have “roles” (leaf, spine) in the network, which simplifies configuration and enhances troubleshooting.
By utilizing the technologies we mentioned before like BGP unnumbered (which simplifies IP address management), EVPN over VXLAN (which simplifies VXLAN management and enables multi-tenancy) and workflow automation (which enables DevOps tools), Cumulus is leading the way to make L3 networking a no-brainer.
Cumulus allows an organization to scale out not only the network, but also the operations team. With Cumulus, the ops team doesn’t need to grow as the network grows. Legacy network tool sets bind admins to typically only managing up to 20 or 30 switches. With Cumulus, your network teams can manage switches at the same ratio server admins do — a roughly 1:500 admin-to-switch ratio. This is the result of reduced network complexities, standardized tools and programs, and automated issue detection. By making these simple changes, you’ll improve time to market and build a better network overall.
These improvements make it easier for an organization to scale their operations efficiently and affordably. In fact, we have found most organizations are saving up to 60% in their total cost of ownership by using Cumulus to increase efficiencies in operations.
When you are ready to look into Cumulus more, we’ve made it easy to try it for yourself. These features we’ve discussed can be tested for free in a virtual private sandbox environment accessed from your web browser with Cumulus in the Cloud. Or, use Cumulus VX, a free virtual appliance that lets you learn and test the latest technology from Cumulus at your own pace, and within your own environment — at no cost. Either way, we enable you to learn open networking concepts, prototype network operations and develop applications risk-free.