Arista, Uber, Silicon Valley

Last week, at least, to me was perfect illustration of how and what media perceives as technology. Everywhere you looked, you saw coverage of Uber and its ability to raise money at a jaw-dropping valuation ($1.4 billion at a valuation of $18.4 billion) and on the flipside was the miserly amount of attention accorded to Arista Networks, an old fashion, honest-to-god technology company that took no money* from venture capitalists and was co-founded by one of living legends of Silicon Valley that went public earlier this week.

Arista (ANET) makes high speed switches that are key to making data centers, the hub of modern Internet work at blazing fast speeds. Arista’s high speed switches don’t make an appearance in Michael Lewis’s Flash Boys, but they should, for those high-speed trading jockeys appreciated the company before everyone else. Microsoft loves it so much that it now makes up about ten percent of their revenue. Arista which has revenues of $361 million and profits of over $42 million raised $226 million from the market ended its first trading day being valued at $3.56 billion.

Every industry has folks who have special kind of genius, some operational, some creative and some technical. Harvey Weinstein and Robert DeNiro come to mind when I think of Hollywood. Ron Dennis is a giant of F-1 racing. Andy Bechtolsheim is that special person for Silicon Valley— a man whose list of accomplishments are longer than most. Many of his peers are playing golf, but he is still starting companies. He has helped start Sun Microsystems, was mucking around with what became routers while at Stanford, started and sold many companies and oh, by the way was the first guy to cut a check for Google guys.

There is a handful of people I have been in awe of — Andy is one of them. It is the greatest dividend of my writer life that I got lessons in networking from him. At Arista Networks, his cofounders Ken Duda and David Cheritorn (who has left the company and his startup, OptumSoft is suing Arista ) are no slouches either. And in a world starved with female leader/rolemodels, there is a Jayshree Ullal. And while, she recently made the wrong kind of headlines a few weeks back, Ullal, a former Cisco executive remains an under appreciated executive.

Arista, from a story telling standpoint is a rich vein to mine. It’s success should be a moment to celebrate that it is an old school tech company. It has invented a new technology, its founders are quirky and eschew the current Silicon Valley system and they live in the far corner of the Bay Area (Santa Clara) which for me is the heart of real Silicon Valley.

So why such as schism in the attention accorded to this company versus Uber (which I love and think they are akin to Google)? The reason — is that Arista’s technology is unseen and also understood by very few, including those in the tech media. Uber, on the other hand is on the first screen of everyone’s phone and they are part of everyone’s regular flow. Tech media corps are daily users of Uber and thus remain fascinated by the company.

For someone to write about Uber, it doesn’t take a lot of effort. It is fairly easy to understand — at least at the surface level — and its firebrand CEO Travis Kalnick is blunt and direct, often saying politically incorrect things that would make even Bill Maher blanch. This flywheel of attention has helped Uber become a verb faster than any other company which in turn has helped the company get money from investors and expand at break neck speed to 130 countriescities in 37 countries.
20140605 VC deals in IT Hardware This divide in software (+services) and hardware is reflected in the declining investments in what an analyst for VC research firm calls, “in the future.” In a report this week, Pitchbook comes up with various reasons why we have seen a major decline in non-software IT companies. Here are the salient arguments from the Pitchbook piece:

  • The proliferation of seed financings has made it easier for startups to receive initial funding and easier for VCs to make low-risk bets on relatively non-inventive companies.
  • Software startups, making apps or social networking startups are generally less-complicated and cheaper to get off the ground, and much easier (and quicker) to sell to potential acquirers down the road.
  • VC funds simply do better in categories where “the innovation cycle is short, such as media and software.”
  • Hardware IT and biotechnology, because the time frame to exit for these types of companies tends to run longer than the typical lifecycle of a VC fund.


While there is a lot of truth in what Pitchbook is saying, but I do think when it comes to hardware-and-software, everything is a lot more nuanced. Back in the day, like Arista, Juniper came up with a new hardware architecture for its routers optimized for the faster and faster Internet. They made their own chips and their own hardware and they also wrote the software that made all that hardware oomph work. The software they developed then JunOS is what has kept them in business. Arista too is cut from the same cloth. Arista has its own software which is what makes it a must buy for web infrastructure giants such as Microsoft. They sell hardware, but in a sense they are a software company.

Similarly, Google which is primarily an ad-and-search company, has pushed the envelope more on core networking and infrastructure technologies than any so called real hardware company. Facebook which primarily makes a social environment and managed a web ID-system has developed hardware to meet its own unique needs. I bring this up, is because the worlds of software and hardware are more entwined than we think.

However, the gulf is that of perception. In order to understand Arista, you need to understand the networks and how they have evolved. Uber on the other hand is easy to grok — and fun. I mean who doesn’t like to use math to pick apart a company blog post. (Oh, hi Felix!) From an investor standpoint, it took Arista Networks ten years to get to public markets. Its returns languish behind, say a company like Uber which is worth about 9 times without even tapping the public market after five years of being in business. So, if you are a professional investor it makes sense to take the shorter route to profits. 

That explains why we are seeing fewer investments in the older style of tech companies, which in turn has forced the hand of companies like Facebook, Google, Apple and others to look inwards to find ways to push the envelope

Arista and Uber, two very successful companies do a good job of illustrating the dichotomy of technology industry and the increasing gulf between the classic Silicon Valley and the new San Francisco-centric technology ecosystem that is more focused on creating tech-enabled services and new kind of media.

Update: *Khosla Ventures is a small investor in Arista Networks. Vinod Khosla was one of the co-founders of Sun Microsystems.


  1. Noel Bilodeau (@NoelCBilodeau) says:

    June 10th, 2014 at 8:48 am Reply

    Great blog post. There’s a lot involved behind the scenes to make our always-connected world possible. Companies like JDSU contribute a lot of the software, hardware and services that keep network infrastructures humming along. If we do our (silent) jobs right, that ‘always on’ capability allows for cool innovations like Uber to happen. Noel – JDSU

  2. Chris Yeh says:

    June 9th, 2014 at 2:36 pm Reply

    I’m glad to see someone celebrating yet another chapter in Andy’s amazing career. Here’s a guy who would be a legend just from the first 5 years of his career, yet he’s still building billion-dollar companies.

    I had the great privilege of sitting next to Andy on the bus ride to the SVForum Visionary Awards. I had no idea who he was, or that he was one of the honorees that night. He couldn’t have been nicer, and was honestly interested in what a young punk like me had to say about the Internet.

  3. Vijay Thirumalai (@vijayth) says:

    June 9th, 2014 at 1:23 pm Reply

    Om – Great article as always. I invite you to consider the current tech landscape from a different vantage point. New industries ( Internet tech in general) and new marketplaces share a lot of similarities with development of cities and nations. The first things to come up are – mostly border outposts (for the internet: core routers, a few telecom companies managing access) , then come the network of roads and bridges (for the internet: connection infrastructure, T1/T3/OC access lines, some basic routers), then come faster super highways, airports, toll roads (for the internet: giga bit routers, optic fiber access, etc..etc). When much of the bigger infrastructure is completed, the focus(innovation & capital) turns to a more under-served local infrastructure. As these cities grow and support a critical mass of people, other public infrastructure local to the city takes over – much like apartment complexes, storage places, malls, restaurants (for the internet: shopping sites, online storage, cloud storage, email etc). As these markets become really stable, the focus (innovation & capital) turns to industries that serve the needs of the hyper local populace – services, security, etc..etc (for the internet: consumer tech, health tech, edu tech)..

    In some ways, we are all lucky to see a completely new mega industry take shape and see its evolution. Also, financially there is a lot of value to be captured in the new frontier than the old frontier.

    1. Om Malik says:

      June 9th, 2014 at 1:33 pm Reply

      Thanks for your comment. I very well understand all the issues — and have articulated them at various points in my writing career. The point of the piece is that we have skewed attention/focus from media on what it seems as technology instead of taking a more 360 view of the ecosystem. In not doing so, we lose the way.

      To think it another way, the fact is without core improvements in the boring infrastructure technologies our cities would be a mess, and the shiny malls and apartment complexes are going to be pretty pointless. In focusing the spotlight on all aspects of the development means that at any given time you view the big picture and make decisions.

      1. Vijay Thirumalai (@vijayth) says:

        June 9th, 2014 at 8:10 pm Reply

        Yes. Completely agree. Point well taken.

  4. ksrikrishna says:

    June 9th, 2014 at 10:03 am Reply

    Reblogged this on Design of Business and commented:
    If as @om says this is bad in the Valley, its 100x worse in India. Our definition of a tech startup leaves much to be desired.

  5. coleinthecloud says:

    June 9th, 2014 at 7:12 am Reply

    ED of Open Compute here.

    When cloud companies actually figure out how to federate. You will suddenly see a huge push for non lock-in commodity hardware. This will bring new speed to the pace of innovation for SV hardware companies. There is a reason Andy is in my board.

  6. pbruklyn says:

    June 8th, 2014 at 7:19 pm Reply

    Sorry, has Uber expanded to 130 countries or 130 cities? If it’s the former THAT is incredible.

    1. Om Malik says:

      June 8th, 2014 at 9:12 pm Reply

      130 cities in 37 countries!

  7. Natalie So (@natalieso) says:

    June 8th, 2014 at 4:43 pm Reply

    Om I loved this post! Such an acute analysis of the tech industry, and you pinpoint some of my feelings about the “hype” that surrounds the newest apps/software that seem to skimming the surface of, rather than advancing, technology (“tech-enabled services,” like you said). Having grown up in Silicon Valley (born in Mountain View), my memories of the classic Silicon Valley that you speak of are actually through my parents’ and parents’ friends’ eyes — most of whom worked as a part of the old guard — my dad worked on printers at Hewlett Packard and my mom ran her own hardware sales company out of a warehouse in Sunnyvale. The kind of expertise required to work on hardware is immense and deep — my dad had a PhD in electrical engineering; and I know friends’ parents who similarly were working on microchips and other hardware systems — there does seem to be a divide in the type of knowledge that is thirsted after in software vs. hardware (the former: SEO, SEM, tech-enabled marketing?). You obviously have seen the tech industry since its birth and have such incredible insights about it — always a pleasure to read and digest your thoughts!

What do you think?

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Enter your email address to follow this blog and receive notifications of new posts by email.


Get every new post delivered to your Inbox.

Join 15,355 other followers

%d bloggers like this: