An afternoon with the inventor of Ethernet – Bob Metcalfe

Earlier this month I had an opportunity to attend a talk by the most well-known co-inventor of Ethernet – Bob Metcalfe. In May, the networking world celebrated the 40th anniversary of the invention of Ethernet at the Computer History Museum, where Metcalfe was honored and invited to speak to employees of the company that he had founded in 1979 – 3Com. Of course, HP acquired 3Com in 2010, so he had really come to HP to talk on the evolution of Ethernet as well as what has kept him busy the past 40 years.

Metcalfe began by stating that the design behind the Ethernet protocol he co-invented in 1973 had changed so significantly that he is often given far more credit than he deserves. Amusingly, however, he said he will not give that credit back. He believes that there is very little in common between the types of Ethernet standards we have today from the IEEE (Gigabit Ethernet, Ten Gigabit Ethernet, 40 Gigabit Ethernet, 100 Gigabit Ethernet) and the original 2.94 Mbps standard that he came up with with the intent of printing 500 dots per inch with a speed of one page per minute. The day of 1 Terabits per second Ethernet is not far off, with the dependency being on the IEEE assessing the availability of components in a timeframe so that devices can be made economically.

Metcalfe spoke of the battles Ethernet had with Token Ring in the early days. Token Ring was heavily backed by IBM, but had rigid standards and was inflexible to the growing needs of the market. Ethernet, on the other hand, was able to spread widely because it constantly adapted, the prime example being the opening up of media support from thick-net coax to twisted pair cable. Ethernet also sought support for higher speeds, soon from 10 Mbps to Fast Ethernet (100 Mbps) by which time Token Ring with it’s support for just 4 Mbps and 16 Mbps was proven obsolete. Another reason Ethernet thrived, he claimed, was that it was not designed to solve every problem. For example, in the ISO hierarchy, Ethernet does not address Security (mainly because he felt it was not appropriate to solve that problem at the hardware level). Of course, now it is standard design to be cross-checking the source address field. He then went into a tangent of how he believes the Internet has an ideological problem in that anonymity is given a high priority, which is a mistake. Metcalfe feels the ability to have anonymity should be an exist, but not as the default.

He talked at length of the pervasiveness of Ethernet into various horizontals. For example, while Ethernet was designed to be a LAN in a building, it has also entered the WAN by killing SONET, an accomplishment he has taken significant pride in. SONET and T1 were both introduced by AT&T, the other ‘big bad corporation of the time’. At 1.544 Mbps, a T1 circuit was half the speed of the original 2.94 Mbps standard, and it was only a matter of time before future Ethernet standards would prevail despite the emergence of X.25, Frame Relay, and ATM. Today, across the WAN, Ethernet is represented as Carrier Ethernet in a $34 billion market of equipment. And of course Ethernet has also manifested itself wirelessly as WiFi.

Today, Bob Metcalfe is a Professor of Innovation at the University of Texas at Austin. Previously he was a Venture Capitalist at Polaris Ventures for 10 years. On the topic of innovation, he also spoke of a few models, including one he coined – Doriot Ecology, named after George Doriot, one of the first modern VCs from Harvard Business School. The premise of the Doriot Ecology are briefly:

  • Startups out of research universities are the most effective at innovating. However, they also depend on funding agencies like NSF and DARPA, research professors, graduating students, scaling entrepreneurs,  strategic partners (such as the large network vendors like HP), and early adopters.
  • Startups need partners to scale. Large companies need to practice open innovation, and be receptive to ideas that come from the outside.

He related a few other models with the way businesses were run in his days at 3Com and Xerox PARC:

  • Intrapreneurship – Here, innovation comes from inside the company. 3Com never had a research division, but tried to push its product groups out to find a prospect. In such cases often the money making groups try to kill off the research group because they don’t generate any revenue and there is pressure at every budget cycle.
  • Spin-in, where efforts to come up with innovation are put up outside the company with the understanding that if it succeeds, it will come back to the parent company. We’ve seen that with Insieme.
  • Spin-outs, where the company has to decide whether it will be hostile to the spin-out or supportive. Metcalfe talked about how Xerox noticed that Adobe, Apple, Sun, 3Com were all exploiting technologies (such as the mouse and the GUI) that were developed at PARC. At that point Xerox started investing in their spin-outs rather than being hostile to them.

Metcalfe said corporate research has deteriorated a lot since his days and should not be reconstituted. While Xerox PARC is now known as PARC, it is nothing like it was forty years ago. For its strength of 25,000 employees, the now defunct Bell Labs ‘only’ had the transistor, Unix, the Princess telephone, and DWDM to show for. (I’m not the first one to note that Bob Metcalfe tends to make controversial statements!) He argued that the only companies that can afford to undertake research (not development, but science) are monopolies. And as was seen at Xerox PARC, monopolies are the least motivated to scale up technologies that they develop. Funds should not be put in corporate research labs or government research labs. In his opinion research should be left up to research universities, such as UT Austin, Berkeley, and Stanford. Professors should be encouraged to start more companies.

He claimed to have a short attention span to re-spin his career every 10 years, from 3Com to Venture Capitalist to Professor. Isn’t it fun going up a learning curve? He finished by saying he recently learned Python in a Massively Online Open Course (MOOC) that he took with his son and 60,000 other students (he got a 90, his son got a 70).


Panel Discussion on Software Defined Infrastructure for OPEN Forum 2013

It has been a long time since I’ve posted a blog entry. In the meantime I have helped put together a SDN panel discussion for OPEN SV this Saturday, June 8th, 2013 at the Santa Clara Convention Center. I wrote a guest blog piece for OPEN SV here with a perspective on SDN and on the speakers.

It promises to be an exciting discussion. Just today, the moderator of the panel, Craig Matsumoto, who is the Editor of Light Reading covered Big Switch’s decision to pull back from its role in OpenDaylight. The panel discussion will also feature the VP of Engineering at Big Switch, Howie Xu. Rounding up the panel are Luis Robles, (VC at Sequoia), Awais Nemat (CEO of PLUMgrid), and Dominic Wilde (VP of Product Management at HP Networking).

If you’re in the San Francisco Bay Area, I highly recommend attending this panel. It is part of OPEN SV’s 10th annual OPEN Forum. More information on the panel itself is available here.

Hello HP. I’m RFC 1925 Compliant.

RFC 1925 – The Twelve Networking Truths is one of the less technical RFCs that are cited by the networking community. It is perhaps best known for the pigs statement, Truth #3:

With sufficient thrust, pigs fly just fine. However, this is not necessarily a good idea. It is hard to be sure where they are going to land, and it could be dangerous sitting under them as they fly overhead.

My favorite, however, is the one that follows it, Truth #4:

Some things in life can never be fully appreciated nor understood unless experienced firsthand. Some things in networking can never be fully understood by someone who neither builds commercial networking equipment nor runs an operational network.

I interpret that as what makes for a successful product manager. Within networking I’ve been a Developer, a Support Engineer, a Consultant, a Network Engineer, a Technical Marketing Manager, and, most recently, a Product Manager for a startup. And now I’m very proud to share the news of my appointment as a Product Manager at Hewlett Packard. I’ll be owning the lifecycle of HP’s Campus products portfolio. HP Networking sponsored a recent Packet Pushers show and is one of the pioneers of commercial SDN solutions. In my new role I’ll be focusing on tackling the challenge of BYOD and VDI in the ever-changing Campus networking environment. Exciting times ahead!

Collaborating with the Networking Community in the Age of Information Overload

Literally speaking, 2012 was the hottest year in recorded history, though there will always be climate change deniers. From the perspective of networking as well, it was a very hot year. Dozens of vendors are battling it out to claim their share of the SDN pie, a market, which IDC expects to grow to $3 billion by 2016. With IAAS/Cloud finally living up to the hype it generated five years ago or so, we are truly in a golden age of innovation in networking. Greg Ferro often says that the last time networking saw such excitement was when MPLS was introduced. However, MPLS was always a Service Provider solution and just a direct replacement for Frame Relay and ATM. If you ran a mid-size Enterprise network or an SMB, the chances are that you wouldn’t need to worry about MPLS. Some have argued that MPLS can be run in the Data Center, but the number of implementations is quite few. More importantly, MPLS had no consideration about the type of applications that it was transporting. SDN, on the other hand, with its Northbound API, is completely application-aware. With all the monumental changes happening in networking nowadays, it can be rather overwhelming trying to keep up just by reading blogs and newsletters. In this post I’ll outline three ways of collaborating with the networking community.

Packet Pushers, which the aforementioned Greg Ferro co-hosts along with Ethan Banks, is the premier podcast show for getting the scoop on trends in the networking industry. It features quality professionals, many of whom maintain their own blogs or are active on Twitter. Packet Pushers has a handy forum where you can ask questions on just about anything and can interact with like-minded networking professionals in the virtual meeting room. Greg and Ethan complement each other very well. While Ethan is more in tune with the more day-to-day activities of a network engineer, Greg is generally more active in promoting the discourse for newer technologies, such as the OpenStack Quantum project. The shows generally tend to be more in favor on Data Centers and SDN than, say, VoIP or Wireless, but thanks to the forum, listeners can chime in with their preferences for upcoming shows.

SDNCentral was launched in January 2012 as means for people to educate themselves on the SDN market and it does a wonderful job at that. One of the website’s features is the SDN Trending Index, which measures the most popular SDN companies, based on SDNCentral community activity. This is a clever way to gauge how hot a new SDN vendor is. A more recent feature of SDNCentral is the Demo Friday series in which an SDN vendor demonstrates their product. At the time this post is published was the second in this series – Cloud-enabled Networking–NEC ProgrammableFlow SDN in Action. The first in the series was Plexxi and Boundary. I had written about Plexxi after listening to them in a sponsored Packet Pushers show. I have since softened my stance on them thanks to the demo, which showcased Plexxi’s optically-connected switches built around a closed, controller-based architecture. I was impressed with how it flattens the network and how it can co-exist with legacy network designs. Indeed, it would be difficult to survive nowadays with a rip-and-replace strategy. From SDNCentral: Boundary applies analytics against real-time network flow data to enable Application Performance Management without the need for appliances or tap/span ports. The demo showed how Boundary discovers real-time application topology and monitors application throughput, latency, packet retransmits and other metrics on a per second basis. In other words, it is Software Defined Monitoring. Without SDNCentral, I probably would not have learned about Boundary or appreciate the value Plexxi can offer.

Ben Pfaff speaking at the Bay Area Network Virtualization Meetup at Hacker Dojo on March 20, 2013

Meetups provide an excellent opportunity to learn by interacting with real people in a face-to-face environment. In the San Francisco Bay Area, there are a few meetups that are bringing a sense of community to the networking industry, fueled by the Open Source movement. It wasn’t like this between 2000 and 2010. Hackathons were traditionally associated with only developers, not networking folks. This week, Nicira’s Ben Pfaff spoke at Hacker Dojo of the past, present, and future of Open vSwitch, which he helped create. He showed a live demonstration of how OVSDB, the configuration tool of OVS, works. I met some of my former colleagues and other peers who I normally interact with online. Nowadays, in the SF Bay OpenStack meetups led by Mirantis and Sean Roberts from Yahoo!, attendees bring their laptops and help each other through the OpenStack installation and configuration process with DevStack. Similarly, the Bay Area Network Virtualization meetup offers a fantastic opportunity not only to learn about OpenFlow and Open vSwitch, but also to mingle with fellow practitioners. However, meetups are not limited to the San Francisco Bay Area. In a recent Packet Pushers show, Kyle Mestery, one of the original team members of the Nexus 1000V, mentioned that an OpenStack meetup has also started in Minnesota. Meetups tend to catch on like wild fire. Hopefully we’ll see many more that cater to open networking.

These are healthy signs of a growing industry with plenty of people willing to help out and give back to the community.

OpenStack Cloud Adoption Picking Up

I was listening to the recent Packet Pushers podcast Show 138. HP, who sponsored the show, spoke about HP Cloud, a public cloud that is built on OpenStack. It was launched in May 2012 and made generally available in December. HP is taking Amazon head on in the cloud provider space (at least when it comes to SLAs) as the GM of HP Cloud Services puts it. HP Cloud comprises multiple availability zones, identity management, account management, block storage (ala AWS S2). Compare that with the Cisco Edition of OpenStack, which is just a validated deployment, tested on Cisco’s UCS servers and Nexus switches only. In other words, Cisco does not offer an Enterprise-grade cloud like HP or Amazon do.

And now you can throw IBM’s name into that hat. This week IBM unveiled a new cloud offering based on open cloud standards, including OpenStack, that significantly speeds and simplifies managing an enterprise-grade cloud. IBM, like HP, is one of the founding members of OpenStack and is launching this OpenStack-based public cloud offering in beta mode about one year after HP did with general availability expected later this year. Amazon is famous for it’s low price structure and it will be a challenge for HP and IBM to compete.

Mirantis, a 13-year old company with a history of strong professional services has immersed itself into OpenStack consulting and training in the last couple of years. They help service providers and enterprises build and deploy OpenStack cloud infrastructure. Some of their customers include NASA, Dell, AT&T, and Cisco. Another customer, Internap, launched its OpenStack-based public cloud service back in October 2011 and was assisted by Mirantis. I installed OpenStack in a personal lab at home and encountered several obstacles along the way. It is not easy to set up. For simple lab installations, DevStack is a good option. However, for production deployments, there is a lot of complexity that needs to be understood and overcome for enterprises to deploy clouds. Companies like Mirantis help bridge that gap. Blog Posts

Musings on How to Build Modern Applications in the Cloud and Onprem