Open Networking: Cisco and Juniper

This is the first in a set of blog posts in which I hope to cover the idea of Open and it’s value to the dominant players in different networking spaces.

(Disclaimer: I am the president of The Network Device Education Foundation, Inc. a non-profit that has projects such as OpenSourceRouting (Quagga) and Router Analysis.  My personal goal is to educate and help the network community, big and small, to build better networks through knowledge.)

Introduction

There has been a lot of discussion on Twitter, Blogs and other social sites (LinkedIn, etc) about the coming Software Defined apocalypse.  Whether the discussion is about software defined networks, data centers, or enablers (such as ODM switches), there is a general picture that the Software Defined products / enablers are trying to destroy dominant player x in the space.  I view things differently: now is the time for the dominant players to focus more on extending their strategy to include Open in order to defend their market share.

Information necessary to explain my point

Lets look at some facts :

1.) Most if it not all network device vendors sell one or more rebranded ODM device.

An example ODM would be Advantech, which has been around for over 30 years and builds ODM appliances sold by many networking companies.

2.) Google was very upfront with networking vendors about their networking needs, yet Google’s needs were not fulfilled, therefore Google and later Facebook looked for ways to provide for their own needs.

My statements on Twitter

Earlier on Twitter I made a few statements

1.) The day an enterprise can buy a Cisco router/switch and put a different NOS on it is the day Cisco gets it.

2.) Cisco, Juniper and VMWare have a few years of dominance left.

3.) By a few I mean a minimum of two years before someone offers a competing product that is well supported and the enterprises can accept.

Screenshot 2015-02-21 19.19.01

 

My goal with the tweets was not to proclaim the death of Cisco/Juniper/VMWare, but to bring attention to the fact that over time Open solutions will catch up and that it is up to the dominant companies to maintain their dominance.  I am going to exclude VMWare from this post as this post focuses on networking hardware and VMWare/Virtualization requires its own post.

My view of the industry need

The value of Open Networking and Openness in general is not just cost, it’s also features, functionality and most of all flexibility.

The Open Networking market is being hit with “me too, but cheaper!” products that can take the place of Cisco/Juniper/Huawei/etc devices.  These products also offer advanced features, but this is not a compelling driver for enterprises at this time as support (i.e. 24x7x365 phone/email/patches) is limited.

Enterprises are looking for flexible devices that provide stable, supported feature sets and allow for future functionality. Companies like Cisco and Juniper have very good support systems, but Open Networking will catch up in the next few years.

Why a few years?

My view is that in order for open networking to take hold in the enterprise the hardware and software needs to be out and tested for at least a year if not two.  While software and hardware exists, and the hardware has been tested (remember ODMs have been around for a while), the software is still young and most of the testing has been done by bespoke operators with specific needs.  Enterprises buy Cisco and Juniper because they trust the brands and the Network Operating Systems are proven.

Current Openness within Cisco and Juniper

I think it’s important to note that over the last two decades, Cisco has done a lot of work to bring alternate operating systems onto their platforms, especially the 6500 (with AYR).  It’s also important to note that Juniper’s JunOS was and still is based on FreeBSD.

*** Update – Mr Joel Krauska informed me on Twitter that AYR also ran on the Cisco 7200, GSR and on PC platforms. ***

Cisco has recently announced ONIE on the Nexus 9000, but does not officially support other NOSes.

Juniper has also embraced Open Networking with the introduction of the OCX1100 and has worked to decouple their hardware and software.

What Can Cisco, Juniper and Others do?

In my mind, it is clear that Open Networking will dominate the enterprise space in the next 2-5 years. This means that Cisco and Juniper need to develop open networking strategies focusing on the value of their software running on what will eventually be commodity switches.  Having a Cisco or Juniper NOS on the Facebook Wedge would provide enterprises with the support they need while allowing the enterprises to feel secure that they have the ability to choose best of breed for their situation.

Cisco and Juniper (and others) can also focus on providing support for other network operating systems on their current hardware, by providing a HAL, allowing their customers to deploy what they need.  As both Cisco and Juniper have some rebranded ODM products, a HAL for those would be a straightforward project.

The value of allowing other NOS on Cisco and Juniper’s current hardware is great and the downside is minor, especially as Cisco and Juniper focus on opening and extending their NOSes.

The Value of The Facebook Wedge and 6-pack Switches

Innovation is core to the Internet and competition is good for consumers.  While this is true,  Facebook is innovating, but not competing, even if Cisco claims they are.

Since the beginning of networking, there have been hacks. No device or feature fits every customer.  Whether these hacks were in the form of scripts, hardware modifications, software features or complete systems, they were born out of need.  Facebook has a need.

I have been in the industry long enough to have seen companies like Blizzard, Akamai, Hotmail and many others conquer scale issues. The initial solutions were ugly (motherboards screwed directly to metal shelves..) but they were brought on by a need that was not being fulfilled by the computer industry.

We are in the same place today, but in the networking space.  Companies like Google, Facebook and Amazon are hampered by the products that are being offered by the big players, Cisco, Juniper, etc.  It’s not just price, it’s also features, functionality and support.

So with that background, it is easy to see why Facebook, Google and smaller companies like NGI “SDN for Real” would build (or design) their own routing/switching hardware.  These companies already did it for their server infrastructure, and a switch, at it’s core, is just a specialized computer.  It wasn’t long ago that many of the main Internet routers were repurposed mini computers with multiple interface cards, such as the ANS RISC System/6000-based T-3 routers.

Another driver helping Google, Facebook and others build switches is the availability of Open Source Networking Software.  While routing stacks such as gated, zebra, etc have existed for a long time (again, back to the repurposed mini computers) the availability of “whitebox” or open switching hardware and features like DPDK from Intel are a more recent occurrence. This open hardware allows people to utilize Open Source Networking Software such as Quagga.

From the support side, one of the benefits of Open Source Software is the ability for companies to modify, patch and tailor the software to their needs.  This is very different from the previous support model where companies would have to go to their vendor, ask for a fix/feature and wait.  Today companies can change it themselves, hire someone to change it or wait for the feature/fix to come in a later release.