November 14, 2016

We’re Launching From Stealth to Combat the Complex and Error-Prone State of Networking

by Charlie Elliott

After years of work, we’re proud to leave stealth today and officially unveil Forward Networks.

Our Background

My three co-founders: Brandon Heller, Nikhil Handigol, Peyman Kazemian and I were fortunate to be members of the team that created Software Defined Networking while working on our PhDs at Stanford University. During our research, we built and operated OpenFlow-based networks from scratch. For these types of networks, very few network software programs and tools were applicable — making the operation, maintenance, and improvement of such bleeding edge networks extremely difficult, and labor intensive. Later, we talked with a variety of network operators, and to our surprise, learned that the pain we had felt running some of the earliest OpenFlow networks, was the same pain felt by those running any type of enterprise network.

Given our experience with Software Defined Networking, we knew that the trajectory for networking was towards larger, faster changing networks. So any pain felt today, was only going to get worse. And left unchecked, such pain would ultimately prevent progress in networking. This was an unacceptable outcome from our point of view.

In 2013 we graduated, and have since assembled a team of the most talented individuals I’ve ever had the privilege of working with. Our team spans the gamut of experience, from creating the largest of distributed systems, operating some of the most complex networks in existence, to building user experiences used daily by thousands of individuals.

 

The Problem

Here are a few characteristics of modern enterprise networks:

  • Diverse - typically containing gear from a variety of vendors, each with their own unique operating system(s) configured and operated differently from all others
  • Large - anywhere from hundreds to tens of thousands of network devices, each of which contains hundreds to over half a million rules governing individual device behavior
  • Fragile - even in well architected, physically redundant networking environments, the network can be one mistaken change away from a major business impacting outage

Current networking software does not address the cognitive burdens placed on network operators by such networks. Thus we see networks that are slow to evolve, susceptible to security vulnerabilities, and frequently fail.

 

The Solution

Solving such major challenges required a correspondingly major breakthrough to improve the state of the art. This came for our co-Founder Peyman, when he published the world’s first paper detailing techniques for how to both accurately model complex network device behavior, and formally verify critical properties of networks. These techniques have been used in other areas where the system is complex, and failure has catastrophic financial or safety impacts such as integrated circuits, communication, software, and aerospace industries. Network outages and breaches can clearly be catastrophic financially, and increasingly impact safety as well.

At Forward Networks, we are drawing on years of experience in this area, including Peyman’s work at Stanford, plus our three years of applied research. We use these techniques to build the world’s most accurate software-copy of your network, using what we have named the Forward Platform. On top of the Forward Platform, we have created three initial applications:

  • Forward Search: Instant access to the entire end-to-end behavior of the network, i.e. searchable access to every possible way the network can forward traffic. Without needing to send packets, you get the full layer 2 through 4 details of what happens at every device along a selected path.
  • Forward Verify: Ensures the correctness of the network’s forwarding behavior and security posture. It instantly uncovers inconsistencies between intent (network policy) and actual operational behavior by validating user-defined high-level policies, also known as “checks.”
  • Forward Predict: Provides comprehensive answers as to how proposed configuration change(s) will affect not only the device on which the change(s) would be applied, but also how change(s) affect the behavior of the entire network environment, including the impact on applications using it - without ever touching your production environment.

These applications eliminate much of the manual, error-prone activity by network operators today, enabling more efficient workflows with degrees of accuracy and correctness never before available in networking.

 

Customer Traction

Today’s announcement follows on the heels of significant customer success. Some of the world’s top companies in the financial, telco, and IT services verticals are already relying on Forward Networks to keep their networks healthy and operated efficiently.

Barry Sheils, Vice President of Global Network Services, Hewlett Packard Enterprise, stated that “The Forward Platform enables us to better understand the end-to-end behavior of our data center networks, resolve issues faster, and proactively prevent network events. The results are improved agility, increased productivity and reduced network incidents.”

"As cloud computing increases in importance there has never been a time where data center network availability and security have been so important," said Peter Christy, Research Director with 451 Research. "There couldn't be a better time for much needed advanced "big data" and analytics platform like Forward Networks to be brought to market."

 

What’s Next for Forward Networks

The validation and recognition exemplified by our customer success and upcoming pipeline is amazing. We’re thrilled that investors and customers alike believe so strongly in our product and vision. As we leave stealth today, we’ll continue to drive customer traction, improve our product, grow our business and shake up what has traditionally been a sluggish industry.

Stay tuned for exciting updates from Forward Networks — you’ll be hearing a lot more from us soon!

Subscribe to our blog!

RELATED FORWARD CONTENT 
November 10, 2022
Why You Should Care About Vendor Hack Weeks

Hack weeks and hack-a-thons are like foosball tables; if you don’t have them, are you even a tech company? These events, once revered for innovation, are now relegated to being blasé and often perceived as little more than playtime for engineers. As someone who’s worked in tech for longer than I care to admit, I […]

Read More
October 13, 2022
How do you know a software rep is lying?

A customer posed this question to me recently; after pausing and smiling (a little too) broadly, he continued, “Their lips are moving.” I thought this would be funnier if it weren’t partly true. The software industry has over-promised and under-delivered for years, making technical executives rightfully skeptical when they hear a new promise. Unfortunately, it’s […]

Read More
September 20, 2022
When is a Digital Twin Entertaining?

When it’s ensuring that tens of thousands of visitors have the best experience possible every single day. Keeping people entertained is a 24/7 endeavor, even the smallest hiccup results in a social media firestorm. Keeping things running requires thousands of dedicated employees and a staggeringly complex network that sprawls the area of a major city […]

Read More

Sign up for our newsletter

crossmenu linkedin facebook pinterest youtube rss twitter instagram facebook-blank rss-blank linkedin-blank pinterest youtube twitter instagram