I met Martin Levy for the first time in Honolulu at the Pacific Telecommunications Council ’2007 conference. After several coffees at the Kalia Tower, and an hour or so discussions on data centers, networks, and IPv6, I knew I had found a true evangelist in the Internet industry. Several more conference coffees in different locations around the world, and I became one of his IPv6 disciples.
As a senior member of the Hurricane Electric team, Martin enthusiastically spreads the IPv6 word to locations around the world including Slovenia, Hong Kong, Amsterdam, Taipei, Brussels, and the European Commission – in addition to acting as a consultant to IPv6 developers and global digital government policy groups.
An accomplished speaker and writer, Martin brings a unique talent effectively delivering IPv6 thought leadership and actual IPv6 network deployment experience to the Internet community.
This is part one of a Pacific-Tier Communications Thought Leadership series interview with Martin Levy, Director of IPv6 Strategy at Hurricane Electric. Hurricane Electric is a leading Internet backbone and colocation provider specializing in colocation, dedicated servers, direct Internet connections and web hosting.
________________________________________________________________
“Internet Protocol version 6 (IPv6) is the next-generation Internet Protocol version designated as the successor to IPv4, the first implementation used in the Internet that is still in dominant use today
” (Wikipedia)
“With only about 10% of IPv4 address space remaining, organizations must adopt IPv6 to support applications that require ongoing availability of contiguous IP addresses.” (ARIN)
“Organizations relying on the Internet to conduct business have only a limited time to act and adapt to changing technology. Those that delay, run the risk their online services may become unavailable to a rapidly growing number of users.” (APNIC)
Pacific-Tier: Tell me a bit of the sense of urgency on (the Internet community) moving to IPv6, and what Hurricane is doing related to the topic?
Martin Levy: Urgency is a word that has been used now for many, many years when it comes to v6. But the reality is, that we have, for every years that has passed, gotten closer to where there are real limitations on the amount of new v4 (IPv4) space that can be added into the market place and added into the existing global Internet.
2010 really marks a time when we have less than two years of available space that can be allocated to the core registries, to the RIRs (regional Internet registries). And as this year and next year go by, we are going to start seeing rules that have never been seen on the global Internet. We are going to see people with requirements to substantiate their use of v4 space in ways that they have never done till this point.
They will see a requirement for documentation, for signatures, sometimes from corporate executives, officers of the company – at least in the US. This will be a whole new world.
If that doesn’t wake people up to the fact that the world is changing, it is unclear what will.
Pacific-Tier: What is Hurricane doing itself to help push this issue along?
Martin Levy: We have always been evangelizing v6, but we’ve been doing it in a way that the users are encouraged to implement v6. In our case “users” means our wholesale providers (Internet service or network providers) that are buying our existing v4 services.
So we have made it easy at the wholesale level to bring on IPv6 connectivity anywhere on our backbone – anywhere globally on our backbone. That, as well as going out into the community and talking about v6 has been a core effort we’ve brought to the table.
It can get better. In some cases we can help a customer understand just how easy it is now, as opposed to five years ago. There really isn’t, for anybody who had bought fairly new hardware any problem enabling v6. There is a set of golden rules to follow from a security point of view. From an operations measurement and monitoring point of view.
But in reality most people can enable v6 themselves and get their feet wet, with great ease. We have spent our time talking with people and convincing them of that fact, quite successfully.
Pacific-Tier: I hear a lot of companies talking about tunneling v6 through and existing v4 network. Is Hurricane running what we would call a “native v6 network” within your backbone?
Martin Levy: Everything on our backbone is 100% native. The core network, all of the Internet peering ports, all of the customer ports, the connections into our data center customers are all what is called “dual-stacked.” In other words they all run native v6, and, if you want to use the term, native v4.
That means that every connection provided is provided as a pure v6 connection. Now, we also provide, because it is needed, “tunnel broker service.” This is a v6 tunneled over v4 service. We’ve been doing this for many years. And there are users, whether they are at home, on a broadband connection in this country or somewhere else in the world, whether they are a software developer working inside a company that needs a v6 connection for software testing… Or whether they are just a home enthusiast, or in some cases it could be a whole university in some foreign country that has no way to get a native v6 connection. They can use the tunnel broker service.
They can use the tunnel broker service with BGP for full routing if they need to, and connect up to the v6 global network though a tunnel connection. In some cases there is no other way to do it.
But the core of the network, every single POP (point of presence or locations), 26 or 28 of them around the world are all configured native v6.
Pacific-Tier: What is your feeling about how your end users, or your actual customers, are using IPv6 in their networks? Is it becoming a fairly mainstream enterprise protocol, or do you have a lot of work to do to teach or provide thought leadership in the market in that area?
Martin Levy: I won’t lie. There’s an awful lot of education that needs to be done, and there’s an awful lot of work that needs to be done – and in some cases even within wholesale or broadband networks. You can break it down into two or three different issues.
The first issue that touches any network is just their outside connectivity. Their core backbone, and links to the outside world, links into providers like ourselves (Hurricane Electric). Those have to be enabled for v6.
And because they are network entry points, that brings up the issue of network security right at the beginning of the day. The interesting thing is, network security for v6 is really identical to v4 – it’s just the syntax that changes.
The addresses are longer, and you have to use colons instead of dots in the addresses. But the theory is always the same. If I deny access to a particular service over v4, I would deny access over v6. The service could be something as simple as SNMP polling of your core router. It could be more complicated like an internal set of web servers.
Any filtering that can be done with v4 can be done with v6.
The second part that needs to be thought about is what part of your network needs to be first seen by the outside world, or in the v6 arena. And it boils down to simple service like DNS for converting names to numbers. Potentially, if you are an enterprise, inbound and outgoing email.
Obviously, your web site. If you are able to bring up your website as v6-enabled, if you are able to bring up certain web services as v6-enabled, you can take those off the list. But even that doesn’t hit the prime point, which broadband and wholesale buyers of IP transit need, and that is IPv6 connectivity to their end users.
In this area are cable MSO, DSL, or wireless network end user environment, they are going to work with all of the protocols and equipment needed to connect to their end customers, and potentially the education of the end customer.
And that is the part that still needs the most amount of work. But luckily for us at Hurricane Electric, we are a wholesale provider. So our issues are really in getting the first stage done, and potentially helping with the second stage. The third stage is left to the customer. And that (the third stage) is the hard part.
But from a wholesale point of view we get our part done, and we know that we can at least enable IPv6 to move and ensure the routing is as solid as it would be in the v4 world.
Pacific-Tier: So do you see new applications, and emerging technologies such as cloud computing, or global distributed cloud computing models that require a lot of addresses to support their VLANs and their internal process – do you see that helping enterprise adjust or have a better sense of urgency on how critical it is to start employing v6 in their networks?
Martin Levy: The story of IPv6 and cloud computing comes up on a regular basis, and it is a real, real requirement. It doesn’t seem to go away, and the two items (IPv6 and cloud computing) seem to be well-connected to each other.
But what’s more interesting as you talk to enterprises is you start hearing a story of “what are you going to do in a world that internally, the complexity of your internal network has started to push the bounds of how you would run an IPv4 network. Clashing private address space, stuff like that.
So we see even outside of cloud computing, where an enormous number of addresses are needed, that in complex enterprises or enterprise back office systems, we see benefits to the very large address spaces being given out. It may not be considered to be a killer application, but it definitely provides a solution far better than can exist in some legacy v4 environments.
Pacific-Tier: Do you have an opinion on the ability of companies such as Verizon Wireless, AT&T Wireless, T-Mobile, as they deploy their LTE and 4G networks. Will that serve as a further catalyst to force companies into the IPv6 world?
Martin Levy: I think the most pleasing part of that is that we are seeing a clear, solid understanding how and why IPv6 and IPv4 must be taken into account within the LTE or next generation wireless world. If you go back and look at very early documents on other wireless structure that have come into the marketplace, they were always very v4-centric.
This has now changed. Now it doesn’t mean that you and I are going to end up throwing away all our 3G, and in some cases 2G hardware, and be forced to go out and buy LTE or 4G hardware and magically get v6. The reality is the back office requirements for those wireless providers still have a lot of work that needs to be done.
Still, the end-user connectivity is being defined with v6 in mind. I have a lot more faith that as of today we’ll see a lot more items like that show up in the market place in a more seamless manner.
Keep in mind that we already see not every, and not so much the popular ones, but we do see certain smart phones in the marketplace that are v6-enabled and applications capable. They are v6 capable over their WiFi connections vs. their 3G connections. But at least it shows the base technology inside smart phones and smart phone products acknowledges why v6 is important.
People may not be using it very much, but that will change.
Pacific-Tier: Where does Hurricane fit in the big picture with IPv6 today? How do you rank with other networks in your category of size, scope, and scale of your IPv6 deployment vs. the rest of the network world?
Martin Levy: Over the last few years the amount of v6 traffic that we have carried has just grown enormously. It has grown by two different measures.
In actual raw bits moved around, while small compared to IPv4, we’re moving a heck of a lot more IPv6 traffic now than we were a year, or maybe two years ago.
The other measure the number of routes, the number of customers, the number of adjacencies, and the number of peering connections with other core backbones we have. We have taken those numbers and eclipsed every other provider, putting us in the number one position globally.
That is a testament to the network engineers, and the dedication the whole company has (to IPv6). And we’ve really done that because v6 is not a side project for us. V6 is not an “add on” to our existing v4 service. V6 is not something we do as a special. It means that every single connection, every customer, every peer, every interconnect on our network, is v4 and v6-enabled.
We keep each protocol on equal footing so we don’t have at any point the thinking that v6 is special. It is part of our DNA, and it is part of our base thinking for everything that we do on the network.
There is nothing more irritating or annoying to a professional soldier than to watch a movie and find technical errors. A haircut that is out of regulation, a misplaced ribbon or medal, errors in weapon nomenclature, or even unit 
designations and locations. A soldier knows within a millisecond when there is a technical error – and it dilutes even the best story line. Telecom and Internet industry-related professionals have the same emotion when terms, equipment, or architectures are mispresented in movies.
Then along comes an author who has either really done his homework well, had great advice, or simply knows his subject matter cold. Once the credibility is firmly established, then there is an uncanny ability to lay a story on top of that technical credibility, and keep even the most critical geek engaged.
Neal Stephenson’s “Cryptonomicon” was the first novel I had read which met this strict criteria. Did a good job, because I spent most of the next year reading everything he ever wrote, and have kept up since with great stories such as “Anathem.” I trust Neal Stephenson, so I am able to freely indulge in his stories without becoming tolerant of an error-prone technical structure to the story.
I like Michael Connelly and Robert Crais because they correctly describe locations around Los Angeles, where I live, and it helps put their stories into context. Did I mention I really like technically accurate stories?
Just when I thought it was safe, and that I would not become addicted to another Techno-SciFi author, I walk past a row of Paperbacks and spy the title “Daemon,” by Daniel Saurez.
“Daemon,” huh?…
OK, for a communicator the word Daemon has a very specific meaning:
A daemon is a computer program that runs in the background, rather than under the direct control of a user. Daemons are usually initiated as background processes. (Wikipedia)
Skeptical, I have the initial thought this would be another silly novel name dropping some lexicon of the Internet in a title to try and suck in unsuspecting readers. I read the reviews, and hold on,… these are not the average reviews writing a couple sentences out of a random word generator. These are real people, and some of them I know! I mean, how often do you see a review from Craig Newmark (Craigslist) or a director of Cybersecurity and Communications Policy writing testimonials?
Guess it was OK to give it a shot, and spend the $10. As I had a week of investment in airplanes and airports to exploit, maybe I would trip into something that was good enough to get me home.
Within the first couple pages a grisly series of murders gets my caveman senses awakened, however the environment in which the murders are committed, is, well, it is technically really accurate and complex enough to keep a geek engaged. I mean, when we start talking about 480 volt power systems, server farms, air conditioning system, biometrics, remote processes – well, it is clear the author has been around the block a few times.
While he quickly goes over my head on topics related to gaming, he attaches the gaming discussion to the underlying infrastructure like a data intelligence to a frequency. And the characters are as equally screwed up in the head as any real life gamer or software engineer I have worked with over the past 30 years. Saurez gets it, is part of it, and has produced a novel that codifies all the sick, twisted fantasies you would expect a systems engineer or software developer to harbor.
Then he ensures there are adequate personalities, education levels, egos, and human emotions t remind us this is not science fiction, it is reality – as we know reality today, adding a bit of creativity to an existing set of intellectual and physical tools. Most of those tools live inside of our known “cloud” of the Internet, but the potential of this creative thinking behind his story line is feasible enough to bring chills to an engineer’s spine.
A Strong Recommendation for your “Geek’s Reading List”
Neal Stephenson and Daniel Saurez are engaging, technically accurate, and tremendously creative authors. Stephenson’s novels are a bit more difficult to read, as he brings his ideas to an abstraction that is a bit above mindless reading. Stephenson almost tests, mocks, or challenges his readers to step back and see the big picture of his story lines. If you read page to page, you miss the point of his books. But still have a lot of fun reading the stories. Sometimes you pick up additional jewels during your third or fourth read through of the books.
Saurez puts it right in your face, and challenges you to discredit his story line. “Go ahead, prove this couldn’t happen today…”
Both are great, and should be required reading for geeks who need to step back from their Cisco manuals and RFC memorization exercises, and actually experience how creative people can apply our existing and emerging technologies to abstractions of thought. Remember, the engineer can design a tool, but only a user can find creative ways to exploit the tool. Engineers can learn a lot from people who apply our visions to solve problems and enable new opportunities. Having recently finished reading Anathem and Daemon, I cannot pass by a router, switch, or server without thinking….
Read the book
John Savageau, Honolulu
Bjarni Thorvardarson is a rare telecom visionary. He thinks on a level of telecoms at an intercontinental level, rather than a national or local level. Hibernia Atlantic is his current project, and with recent news the submarine and terrestrial cable system is now in the global media distribution business, he is shaking up the telecom community. An Icelandic native, he has brought his knowledge and skills to the United States, basing Hibernia Atlantic in Summit, New Jersey.
Pacific-Tier Communications series on Entrepreneurs and Thought Leaders continues with Bjarni Thorvardarson, CEO of Hibernia Atlantic (www.hiberniaatlantic.com)
Pacific-Tier: Bjarni, what’s been happening with Hibernia Atlantic in the past few years, since I had my last opportunity to visit with you in Summit?
Bjarni Thorvardarson:
We’ve actually had a busy couple years – a very busy couple of years.
As you may recall, we started this business by buying a submarine asset that was formerly owned by 360 Networks. We’ve been busy trying to build our terrestrial network to try and connect this submarine cable to anywhere. We no longer refer to ourselves as simply a submarine cable, but rather a capacity provider in the eastern North America region and in Europe. Less than half of our business is now in trans-Atlantic capacity.
Even though that remains our core competence, and the core of our business, in terms of our business it is less than half of our revenue.
So that is part of the transformation that has happened over the past few years.
In terms of revenue, in 2005 we generated about $2 million in revenue, then $7 million, the $18 million the following year, then $28 million, and last year we generated about $38 million dollars.
And now the target for 2010, with our addition of MediaXstream, is about $75 million dollars in revenue. So, we’ve been growing about 40~50&, up to 80 or 90% a year. So you see it’s a very rapid growth. We are riding on a couple things. One is that we are operating in the biggest capacity market in the world, which is the Northeastern part of North America and Europe. And we are focused on a niche sector, which is the big bandwidth market – which is by itself growing about 40% in volume terms a year
And now, since we still have a relatively small market share, we are growing even faster than the other (traditional) markets (players). That’s how we’ve been successfully growing our revenues.
Last year, in 2009, we were confident for the first time, and we were profitable. We are very happy with the way things are going.
Pacific-Tier: That’s excellent! Can you tell me a little about yourself, and how you got into the business of both submarine and terrestrial bandwidth and capacity?
Bjarni Thorvardarson:
Sure. I’ve been in the IT and telco business for about 17 years since I finished my business studies. By education I have a degree in engineering from the University of Wisconsin (At Madison). Later on I added a science degree from the London Business School.
I went into the telecommunications business, and from there into investment banking (around 1998). Then shortly thereafter I started a fund that was investing in telecommunications and IT. That was 1999 into 2001. From there Ken Peterson actually got interested, Ken Peterson was the owner of Columbia Ventures (CVC), got into an investment in telecommunications. He brought a co-investor in with me, and bought the fund eventually. That’s how it came about that I started working for CVC.
And I’ve been investing on their behalf in telecommunications since 2002.
Now one of the investments that we had made was indeed Hibernia Atlantic. That was 2003. Then in 2004 we sold an investment that I was responsible for here in Ireland. Then I took over the responsibility of Hibernia Atlantic. Since 2005 I have been with Hibernia Atlantic, running it for CVC.
That’s how it came about that I’ve been investing and spending my time in telecommunications.
When it came that I was to take on the Hibernia Atlantic project, it was meant to be a 6 month project. We’d see what we could do, fix a couple things, and recruit the right people. It’s one of those things that a 6 month window turned into a sliding 6 month window. And during the next 6 months we ended doing something exciting. That’s what happens when you get interested in what you are doing.
You see the potentiality, you see what can be done, and you kind of stick around – and so its 5 years now. I no longer refer to it as my 6 month project, I now refer to it as my passion. It’s what I do. I’ve been commuting between Ireland and the US now for the better part of 5 years, and relocated to New Jersey where we have the Summit headquarters, or US headquarters a few years ago. So I am pretty heavily enrolled in the Hibernia project.
Pacific-Tier: That’s good. You mentioned earlier the topic of moving from telephony to broadband. Where does Hibernia and your plans fit into what I would call the “globalization of communications,” or the “flattening of the communications architecture…” How do you fit into that model?
Bjarni Thorvardarson:
Hibernia is very much a long-haul provider. We started in the long-haul wholesale capacity business, so provided the infrastructure provider to other service providers – the likes of BT and France Telecom, Cogent and the like… We did a very good business connecting the biggest consumer markets to places like New York, London, Amsterdam, Chicago, connecting those markets and enabling service providers in those markets to connect to different parts of the world.
Our business is really predicated, it is built on the globalization of business, or the globalization and international movement of information. That was the core part of our business model.
Now since then we have moved on to going up the value chain (if you like) to become the service provider to enterprises ourselves, and begin focusing on the finance vertical, which is a very demanding market. They (financial markets) are demanding and expecting low latency circuits between different trading markets and centers. That was a big first step into the enterprise world.
The next step we took was to the media sector, which we did first when we were acquainted with or partnered up with MediaXtreme, investing in MediaXstream a couple years ago. Then finally culminated in the acquisition of MediaXstream last month. So that’s our big step into the media market.
So now Hibernia’s approach to the market is threefold:
- We are still very much the legacy we started, which is the wholesale provider to other service providers and telcos around the world
- Second is the finance sector
- Third is the media sector
But they all are very much relying on the globalization of business and people’s general view of the world. So we have to look and depend on it.
Pacific-Tier: So in a traditional sense submarine and terrestrial long-haul networks relied on SONET or SDH technologies as the basic (communications) protocol. Is Ethernet playing a stronger role in anything Hibernia is doing now?
Bjarni Thorvardarson:
We started providing waves (2.5 or 10 Gigabit) via SDH and SONET as a product to the wholesale sector. For technical reasons including that was the technology Hibernia was built on. And also that was the product the wholesale providers relied on. They need to connect their different POPs (Point of Presence) equipment. A POP in New York, to a POP in London, that equipment relied on and called for SDH/SONET to connect the POPs.
Now as we grow into the enterprise sector, then the guys, the traders, or whoever we are doing business with – they don’t have SONET or SDH equipment. They have Ethernet equipment or equipment that calls for Ethernet protocols. So it is incumbent on us to be able to provide that without cumbersome translation from one protocol to another.
So we have since built Ethernet at the core of our protocols. Now we can offer Ethernet over SONET, which is dedicated Ether net point-to-point. And we also built, using H3C equipment, a product that we can connect customer to and point-to-point to multipoint capacity.
So moving from the SONET/SDH world to the Ethernet world, or switched Ethernet is very much what we are doing. I am right with you there that we are phasing out one world and moving to another one. Even the telco providers are increasingly moving into the Ethernet world. Especially when it comes to building out their ISP or Internet networks.
Pacific-Tier: When you see organizations like the Carrier Ethernet Neutral Exchange (www.cenx.com) and things like that popping up that are basically designing their product on the old bilateral telecommunication company design,… Do you believe that bilateral Ethernet, or that bilateral carrier relationships still have a role, or will companies like Hibernia make many of those old relationships irrelevant?
Bjarni Thorvardarson:
Hibernia, in its traditional sense, is not going to replace bilateral agreements. But bilateral agreements are going to be phased out when it comes to the exchange of Internet traffic, because exchanges are going to replace them. It is like the minutes (telephone settlement) business extremely cumbersome. If you want to build a bilateral relationship with other telecommunications providers you want to exchange traffic with through some of the voice exchanges you can do business in a matter of days.
And that is the same with the exchange of Internet traffic. If you want to do peering on a bilateral basis with companies it takes you years to build up. If you want to do it through an intermediary (such as a public internet exchange Point), clearly it is moving from the bilateral agreements to the exchanges.
Now how does that related to the price a carrier has to pay when going through the exchanges? To transit pricing? Or what have you?
And we can see where these intermediaries are actually charging less and less for the service of being in-between the delivery of the data and the content origination. You can see that in transit pricing, and how transit pricing is continuing to plummet. So I think that we are becoming less reliant on the bilateral agreement. And I firmly believe the opportunity and the necessity of getting more exchanges up and running is important.
And I think the same transition, you can see the same transition when it comes to not only Internet peering, Internet traffic is also the interconnection of Ethernet circuits, the same transition occurred that we saw 50, 60, 70 years ago when it came to voice traffic. If you wanted to make a call from London to New York you had to call an operator in London, and he made a physical cross connect to a long-distance line that terminated in New York.
The operator in New York then made the physical transition to the local tail line to the customer in New York.
That’s very much the same as when you are setting up an Ethernet circuit today. You have to build up a physical cross-connect in New York between the local tail provider and the Hibernia facility, and then in London to the tail provider over there.
With INNs and with proper Ethernet virtual cross-connects which are relying on a virtual exchange, or like exchanges that you are referring to, it’s going to revolutionize the provisioning and setup time of these Ethernet circuits. We can see a leap in that direction over the next couple years.
Pacific-Tier: One other thing I’d like to ask about Hibernia’s role in the Internet and international community in particular. I’ve been spending a lot of time in developing countries myself over the past year or so. So does Hibernia play a larger role, more than just an economic role,.. Do you also have a larger role in supporting the global community to provide a product that will bring global communications, education, entertainment, media, – can you bring that type of thing to another level?
Bjarni Thorvardarson:
I think that when you have a major company, a large company on the global economic scale, then you have to sit back an think about what your global social responsibility is to the world, and what you can contribute to the world. Hibernia is light years away from being at that size, and we can best fulfill our role now by looking for what is our economical role in this world.
Today that role is to provide large scale, high interconnectivity in all the market we operate in at a very attractive price. And by doing that we can contribute to the successful globalization commerce that will facilitate the business which will break down barriers that might prevent doing business, or from offering access to multinational companies.
That’s really what I think is our role in the world, to enable companies and people around the world be operating seamlessly as if they were sitting at two desks right next to each other (companies), and thus taking away the physical barriers of being located thousands of miles apart.
Pacific-Tier: If you look at the ideas, of say Carr’s concept of the “Big Switch” (Nicholas Carr), where telecom companies, and computing companies, and storage companies actually become nothing more than a huge, ubiquitous utility that people expect. Do you agree with that idea, or do you believe companies like Hibernia should be able to offer much higher value than the idea of a utility “big fat pipe?”
Bjarni Thorvardarson:
Well I think everyone has to know which business they are in, and that people can be in more than one business at a time. I say that from experience, because CVC (his prior venture capital company) was in the manufacturing of aluminum, and the manufacture of advanced products that were made from aluminum.
Making or smelting aluminum is very much a commodity business. The success of the business is predicated, or based on you operating a business efficiently. It’s about cost, cost, and cost. If the price per pound of aluminum that you smelt is higher than your competitors, you are out of the market. So that’s how we operated in the aluminum market.
But we also had exclusion companies. That is changing the aluminum ingot to bars that can be used by manufacturers, and be converted to door frames, and window frames, and converted into materials that could be sold to the end users or consumers.
So we were very much aware of the different needs of the value-added service market and the commodity market.
And I think the telecom business is very much the same. You have to know whether you are in the commodity market, the utility market, and there is a fair amount of utility market in the telecommunications world. I think the core of what Hibernia does is just that. It is a utility capacity between the POPs. It is providing 10G (Gigabit) capacity between London and New York, or London and Amsterdam, connecting all these high capacity markets, and it is a utility market.
You have to be very efficient in terms of how you operate your market.
Then, when you go up to the media market, or to the finance market, it is no longer a commodity market. The trader that is trading between London and New York, he does care about the price he is paying, but even more concerned with no having a second-rate service.
So you have to know which market you are operating in, and telecommunications will remain within the two markets.
Now Nicholas Carr’s concept or theory of the “Big Switch” where the world is going to cloud computing as a utility, where you plug into a socket in the wall and you are connected to a network of computing power is a noble one, and a very interesting one, and I think it certainly is going in that direction, but the difference between the bits and bytes, and the electrons that flow on the wires of the utility companies or the electric companies – it is bits and bytes of sensitive information that you do not want leaving the company or be flowing on the wires outside of the company.
So there are many challenges the “Big Switch” theory or concept. But there are a number of companies that are building up a very successful business model. Amazon being one, and a number of other companies offering cloud computing and growing extremely fast.
I am fascinated by the concept and the model of business, but I don’t think there is quite the pure cut between computing and the traditional utilities.
Pacific-Tier: Any other vision or looks into the future Hibernia may be able to share as you peer into the next 3 to 5 years?
Bjarni Thorvardarson:
I wish I could pretend to have a crystal ball, and say what our visions are, but our vision, really for the near to mid-term future is to continue our growth into the different enterprise verticals. We need to continue to service the market we comfortably define as our core markets, being North America and Europe. That’s where we will continue to focus our attention.
But we will to some extent continue to introduce new products that will leverage our network, and continue focusing on different verticals that we can also continue leveraging the network. The game for Hibernia over the next couple years is leveraging the asset. Those assets are not only our network, but also the experience of our company (employees) – the people we have, the processes, and the systems we have. Our competence and the network – that is what Hibernia is going to be not only for the next few years, but beyond.
Pacific-Tier: Do you see any potential partnerships or expansion across other parts of North America or into Asia at some time in the future?
Bjarni Thorvardarson:
Without any doubt, I am sure we will find partnerships that will benefit both parties, but it is nothing I can speak about or speculate about right now.
Pacific-Tier: Fair enough! Any final words on Hibernia, yourself personally, or what you see as interesting things happening in the market?
Bjarni Thorvardarson:
Not really, I am just enjoying working in this space, and I’m looking at a number of exciting opportunities. M&A opportunities, growth opportunities, and I am just excited to be here.
Pacific-Tier: As we all are, and thank you very much for your thoughts – it has been a great discussion.
Bjarni Thorvardarson:
Thank you! It is my pleasure!
| Mr. Bjarni Thorvardarson is the CEO of Hibernia Atlantic since January 2005. Mr. Thorvardarson joined CVC, Hibernia’s parent company, in 2002 from ISB bank where he launched and managed the listed Talenta-Technology fund which focused on emerging communication and IT opportunities. Prior work experience includes investment banking at FBA bank, management of an MIS department and European Sales Director of an IT company. Mr. Thorvardarson holds an M.Sc. degree in Engineering from UW-Madison, an MBA from ISG in Paris and an M.Sc. in Finance from London Business School. Mr. Thorvardarson serves on the board of One Communications, Magnet Networks. |
Read the Pacific-Tier series on Entrepreneurs and Innovators
The institutional horror stories continue, the old Internet Protocol Version 4 (IPv4) address space is nearly gone, and if we do not transition to IPv6 with its nearly unlimited address space the Internet will grind to a halt.
A recent survey in Europe by the European Commission concludes that even in technology-progressive European countries “few companies are prepared for the switch from the current naming protocol, IPv4, to the new regime (protocol), IPv6.” ARIN (the US-based Internet Registry) agrees, reminding us that “with less than 15% of IPv4 address space remaining, ARIN is now compelled to advise the Internet community that migration to IPv6 is necessary for any applications that require ongoing availability of contiguous IP number resources.”
OK, so what the heck? Why aren’t we listening to those who understand the sense of urgency to migrate to IPv6, and get moving towards establishing a solid migration plan? Are the vendors ignoring the problem, reticent in providing IPv6 support in either application software or hardware, and preventing us from adopting IPv6? Are we just comfortable in our use of IPv4, network address translation/NAT, and are information technology professionals simply afraid to make a stand with management to start making the move?
Most mainstream software providers appear to be making the effort to go to IPv6. Microsoft has IPv6 as an inherent part of Windows and new Windows applications, Apple – ditto. Google engineers Lorenzo Colitti and Erik Kline recently received the Itojun Award from the Internet Society for “contributions to the development and deployment of IPv6.”
All major switching, routing, and server hardware companies are producing operating systems which include IPv6 compliance. Even cloud computing vendors such as 3tera are providing native IPv6 support within their platform and infrastructure as a service support.
What I Want from IPv6
I want everything from IPv6. Everything that has an electronic, communications, mobility, or interface should be addressable. I love the idea the California Highway Patrol can work with a company such as On Star to shut down a stolen car on the freeway before the driver kills himself or an innocent motorist. I love the idea I can work with an electrical utility to provide smart GRID technology to my entire home electrical system and not only reduce my bill, but also lower my carbon footprint. I love the idea I can control nearly anything I own or manage through a smart phone handset.
The IPv6 address space is large enough that we will have more than sufficient means to address everything we want – while smart people start working on IPv10 or whatever is needed for a couple generations down the road. So they can extend IPv10 to the rest of the galaxy.
Of course, unless we move forward and accept the temporary pain of moving to IPv6, none of this is likely to happen outside of some private implementations such as Verizon Wireless’ LTE network. Verizon is forcing the IPv6 issue with handset and device vendors by demanding their “…device shall support IPv6. The device may support IPv4. IPv6 and IPv4 support shall be per the 3GPP Release 8 Specifications (March 2009)” Kudos to Verizon for taking a stand on IPv6.
I further encourage moving my identity to an IPv6 address. Who needs a social security number, =social insurance number, or other identity when I can have my own personal IPv6 address? No identity fraud, as it can be linked to my DNA or other funky unique security code. My IP address, with my DNA and fingerprint, and I have the basic elements of a base for all other communications and identifications. Or I will become a borg.
But I would like to log into my home, and have heat turned on 5 minutes from arrival, the oven warming, favorite TV dinner selected for cooking, and even my 2 liter bottle of diet soda positioned for easy removal. My television set was remotely programmed, and the MP3 player auto-filled with music and other stuff from its docking station to give me something to listen to during my evening run along the beach or Wildwood Canyon.
Every device for my personal life that has a pulse can have an IPv6 address, controllable by me for whatever reason I choose. No IPv6 address for my jog though, as I want to de-couple some things important to life.
Who Cares About IPv6?
Martin Levy, from Hurricane Electric (a global Internet service and network provider based in Fremont, California), is a tireless evangelist for IPv6. A member of nearly every IPv6 working group (real working groups, not social working groups!), Martin travels the globe teaching, chiding, and inspiring networks to make the move. Martin recently recorded an interview with the European Internet Registry/RIPE where he explains His position on IPv6, his company’s approach to IPv6, and reminding the Internet community of the risks of not making the move to IPv6.
Martin strongly advises ” If you’re getting connectivity in a data center as a transit over an international connection, as a cross connect inside a telecom hotel, if you are an enterprise, IPv6 (deployment) should just be a tick mark…”
Martin travels the world in his quest to inform, and encourage those who do accept their responsibilities and urgencies embracing IPv6, such as at a recent conference in Slovenia, where Martin congratulated the Internet networking and content community by stating “Slovenia’s IPv6 initiative has been very successful and is becoming a blue-print for IPv6 initiatives in other countries worldwide.”
Internode, a large Internet network provider in Australia, has joined the movement towards IPv6. Partially because it is the right thing to do, partially because it is nearly impossible to get additional IPv4 address space from the Asian Internet registry, APNIC (Asia-Pacific Network Information Center).
“Our objective is to ensure that Internode has the most experience of any Australian broadband provider with the operation and support of native IPv6,” Internode managing director Simon Hackett said in a statement. “By the time IPv6 becomes a necessary part of connecting new users to the Internet, Internode will offer the very best ‘production’ IPv6 service available in Australia. At that point, for all customers, IPv6 will ‘just work’.” (Network World, 6 Nov 09)
Our Call to Action
Every blog entry is supposed to include a pithy call to action. In this case the call to action is real. We need to adopt IPv6. Excuses will not bring our global Internet-connected and Internet-enabled world together, and will not enable our next generation of network users to fully execute on the promise of exploiting life in the “matrix.”
IT Managers – you need to get off your backsides, and learn, learn, learn, everything you can about Ipv6. It is mission-critical. Then you need to brief your management – the CFOs, CTOs, CEOs, CXOs, and let them know the urgency of re-stacking your organization to accommodate and drive Ipv6.
Networks – If you are an Internet network provider, and you do not support Ipv6, please get out of the business. With all due respect, you are the problem.
Content providers, application service providers, SaaS providers, equipment vendors, and everybody else hanging an Internet shingle on your door. Ditto – if you are not building IPv6 support into your product, you are the problem. Make it easy for the IT managers, individuals, and future generations by taking Verizon’s approach. “If you do not include IPv6 support in your product, we will not use it.”
What is your IPv6 message?
John Savageau, Long Beach
The Cloud Computing Conference and Expo in Santa Clara has come to an end, leaving a fair share of opinion, skepticism, and robust 
discussion for the period of incubation leading up to the next conference. Many companies have adopted “Cloud-something or other” as their new name, and are aggressively bringing their products to market. We observed exhibitors displaying cloud network management software, cloud email and SMS messaging, cloud security – basically cloud everything.
According to the conference organizer (SYS-CON’s Jeremy Geelan), the conference exceeded all their expectations. Those expectations were to enlighten attendees on the state of cloud computing, train those who need to need to know more about clouds, and offer a forum to debate both the value and future of cloud computing.
I try not to be one of those expo attendees who find joy in tormenting booth staff, as often you will meet sales and marketing people who have no technical knowledge of their product, and are trying to simply collect cards and develop potential business from visitors and conference attendees. A model used over the past 100 years of industry conferences and expos.
However, of around 50 booths at the conference, only a handful of vendors could really be considered directly involved in development of cloud products or services. The rest either resell somebody else’s service, have tagged their legacy product with a “cloud” prefix of suffix, or simply set up a booth to have visibility or presence at the conference.
Now this is not a bad thing, in my opinion. As a cloud evangelist, a person who has dreamed about cloud computing, GRID computing, high performance computing, and network computing for most of my professional life, seeing the tag “cloud” plastered on just about everything makes me happy. It means the term and enthusiasm for cloud computing is no longer the domain of engineers, but is about to hit the “hype scale” that will drive the vision into the eyes and minds of just about everybody on the street. Without hitting this phase of “hype” development, cloud will risk dying or fading away like many of the other great ideas of our generation.
If that is what it takes to continue forcing companies to build faster, cheaper, and more agile cloud products; if that is what it takes to push governments to understand the value of virtualization and consolidation; if that is what it takes to push entertainment, social media, the financial community, and all industry information technology planning over to the cloud, then I will gladly buy the lipstick and distribute it freely to marketing companies to smear on the next startup’s branding plan.
Getting Past the “Geek”
As with all well-attended conferences, the most robust discussions took place in hallways, the exposition floor, and café tabletops. It is exciting to be in the early stages of technology shifts, as everybody has a different vision, different direction, and different opinion on the best way to create technologies, and apply them to business and social problems. Bringing back the Internet analogy of the 90′s, when email was considered a tools for geeks (circa 1992), and would never replace robust and mature technologies such as fax, cloud computing has a fair share of skeptics and “nay-sayers” as well.
Why? As engineers we are probably much more agile when jumping on the technology “first-mover bandwagon.” We are the ones with home entertainment systems which frequently pop circuit breakers, and occasionally attract local police departments to gently remind us we are being obnoxious and disturbing our neighbors.
Financial officers, operations staff, sales people, and other professionals are inherently reluctant to refresh technology and processes which work. To disrupt a business process requires a very compelling argument outlining and presenting the need for change, the risk of not making a recommended change, the potential outcome if the change fails, and the “pain point” technology refresh will solve when adopted.
Another example. Today most sales organizations have adopted some kind of CRM (customer relationship management) platform. It might be a SaaS product such as SalesForce.Com, Microsoft Dynamics, or other internal application. 15 years ago no sales person would willingly put their sales “funnel” into an online system, nor would they even give up their address book without a fight. Time has proven CRM systems are good for both the sales person, as well as the company, and facilitates the book-to-bank process. But it took a very long time to prove to both companies and sales people this process was valid, and still today has a strong lobby of reluctant old-timers who resist CRM.
Virtualizing IT applications and consolidating data centers makes sense. Economic, environmental, and performance sense. Let’s support the marketing efforts to bring cloud to the headlines. As engineers we need to be tolerant of those efforts, and understand without the marketing phase of cloud development it will take us longer to get into the DNA of future network and compute technologies.
And solve the final question, “which shade of lipstick is best for the cloud?”
John Savageau, Long Beach, from Sunnyvale, California
So you have 20 years of engineering and business management experience under your belt. It is logical that you will have a level of tacit knowledge many companies will pay for.
In the years following the Dot COM meltdown I remember encountering many consultants. Guys we’d worked with for many years, and knew the telecommunications business cold. Everybody thought when the Dot COM 
meltdown occurred, these guys would do great with their new consulting careers. I mean, with that amount of experience young companies should be throwing money at these guys with all the experience.
In a recent post on a cloud computing mailing list, an old timer from the telecom days admitted that for him “being an independent consultant means about the same thing as “being unemployed.”
That’s impossible. The old guys are needed, and young companies cannot survive without their experience and leadership. Right?
Or is it also possible that technology is changing so fast right now, that like the old telecom guys who could quote the process of determining minutes ratios, settlement, allocators, and all those other tremendously complicated telecom negotiations just fell by the side as really smart younger guys who were tremendously Internet-savvy blew past them?
Yup – that’s pretty much what happened.
Over the past few years the joke was “you cannot swing a dead cat over your head in downtown LA without whacking an unemployed telecom engineer.” Guys that are struggling every day trying to figure out how to squeeze minutes out of Internet phones. Guys who are trying to grasp the idea of losing everything they understood and worked for in the good old days of Nortel DMS 250/300s and AT&T 4E/5ESS telephone gateway switches.
Guess what… it is happening again. The current economic meltdown, and huge unemployment numbers are hitting the telecom industry. Lots of newly established independent consulting companies on the street, established with a couple hundred dollars and a do-it-yourself LLC incorporation website. The old guys are trying desperately to throw together snappy websites with basic static HTML pages that discuss their ability to provide “powerful solutions” to “all your business needs.”
It is almost painful to see. While at the same time energetic young serial entrepreneurs aggressively hammer away at the emerging cloud computing industry, entertainment and content delivery industries, social media sites, and Internet service providers. Ooops, well maybe the guys running the ISPs should start looking at taking a couple extra courses on cloud computing at their local technical schools, as the ISP business may also be providing us with a new community of consultants during economic difficulties…
There is a need for consultants. Developing countries still lack the basic knowledge and skills needed to bring their country infrastructures up to at least “entry-level” global speed. Smaller cities in the US and Canada are still under served by both telecom/Internet services, and local adoption of Internet-enabled services into the business and local communities.
LA does not need consultants. New York, Northern Virginia, the Bay Area – all those places have passed the late stage worker generation, and left them behind in a heap of wandering consultants. Consultants who were CTOs, CEOs, VPs, …KEY MOUSEs.
The older generation (those of us past 35) need to stop, breathe, and assess where we are with our industries and professional communities. We can either go for the next generation of technology with reckless abandon (cloud computing….), or eventually find our way to the telecom “fossil” bar to talk about the good old days and whine about the indignity of our current situation.
Independent consultants are the same thing as being unemployed… Damn…
I won’t stand for that… Watch out cloud computing community!
John Savageau, Long Beach
Independent Telecom and Internet Consultant
“Waste is Good”
So says the Gordon Gecko of data tech, Chris Anderson, in his July 2009 Wired Article of the same title. His article goes on to state “Technology is becoming too cheap to meter. So stop metering. It’s time to harness the power of abundance.”
Using examples such as the Tsunami of junk videos on YouTube (“Making the World Safe for Cat Videos”), Anderson presents analogies from nature to make his case for promoting data waste. If you look at the number of fertilized fish eggs produced .vs the number of fish actually making it to the end of their life cycle, you will see the ratio of survival is almost too small to calculate. One in a million fish eggs actually finds its way to being a fish.
So why bother worrying about waste? Darwin says it is the survival of the fittest, so let DNA develop from the data and file survivors that will make the species stronger.
Thus goes Anderson’s case for supporting waste data. If you can go to Amazon and search the cost and size of hard drives, you can now get 1Terabyte network attached drives for under $125. A terabyte of storage for $125…. So why not back up every piece of data you have ever created, downloaded, or spawned through runaway rogue code?
I remember all too well my own experiences trying to sift through files and files of data, only to try and remove enough junk to get around “disk full” errors. The time I spent cleaning and maintaining hard disks, files of floppies (for those of you who do not know what a floppy disk is, check Wikipedia… Old age is not pleasant), and suffering through ensuring I could travel with enough capacity on my computer… Those days are over.
With my 500 Gigabyte drive at home, I can backup pretty much everything I have ever recorded to a disk. With Multi-Terabyte home system drives popping up, I will hopefully be able to continue filling drives with whatever junk I need for a long time into the future. I will never use 99.99% of the files on the drives, but who cares? As Anderson so eloquently states, “we have a very developed sense of the morality of waste.” But, on the other hand, “nature wastes life in the search of better life.”
Thus of those terabytes of wasted disk space that cost less than $125 per whack, possibly 100 megabytes contain that gem of data I may need at some time 15 years in the future. We don’t know.
Associated Content states there may be in excess of 100 million videos available on YouTube. Nielsen claims there may be more than 5.5 billion YouTube videos downloaded a month. If we wanted to view YouTube content 24 hours a day, we would probably never even approach viewing 1% of the available video during the course of an adult life.
I s this all wasted disk and network space? For me, yes, for the other 2 billion Internet users around the world – well, I am sure there are a large number of people who actually do spend their entire day watching cat videos.
John Savageau, Long Beach
January 2003.
“Hey Adil, I need some help getting a LAN installation done – you up for a month or so worth of consultant gig?”
“Sure, where is the job, and are there any special problems?”
“Well, it is in Mongolia…”
“Mongolia? Are there any real issues with the customer?”
“Well, it is at a new gold mining operation. Location is about 300km from the nearest city, no electricity, no telecom infrastructure in place, and the temperature gets down to about -45c at night. All they need from you to design and implement a fiber optic LAN system within the 150 sqkm campus, and then build a VoIP architecture to bring their communications back to Canada and the capital city (Ulaanbaatar) of Mongolia. Guess you will have to use VSAT (satellite) to someplace like HongKong or California to make the connections.”
“Cool, when would you like me to be there?”
Adil Mehmood is what real engineers aspire to become. With the tacit knowledge gained from more than 20 years in the telecom engineering and operations business, there is literally no job too large or difficult for him to engage. He has specialized in implementing telecom systems and basic telecom infrastructure in developing countries throughout his career – one of those unknown professionals who actually have the privilege of going to sleep at night knowing he has made a huge, positive impact on the future of millions of people.
Part of the Internet tech community hangs out at conferences and parties, others roll up their sleeves and apply their energy and experience to real projects, in countries and locations most of us may not even be able to find on a map. Adil Mehomood is one of those people, and unsung hero of the Internet community.
I recently caught up with Adil as he was passing through Los Angeles on his way back to Mongolia, Kyrgyzstan, or wherever his trail currently heads.
John – Adil, what are you doing these days?
Adil – Well, after spending time back in the UK trying to settle down into the domestic life with my family, I quickly realized I was getting bored . I wanted to be out dealing with challenges, and this meant working as an expatriate again in a developing country. Over the past few years I worked on a large rural VoIP project in Mongolia (first of it’s type) and just ended up staying in North Asia!
John – You are known in the telecom community as one of the more creative network design engineers. How did you get into that level of engineering?
Adil – I think one of my inspiring moments was in 1995, I had just landed in Beijing on a look-see trip for a 2-year contract with Sprint China. I ended up at the Beijing Telecom data centre the same afternoon, helping some engineers from the Beijing Telecom Authority to upgrade their initial Internet connection to 256Kbps. This was back in the days ChinaNet had only two 64Kbps satellite links to California serving the entire public Internet in China. I never looked back, and ended up working with some really talented folks in Beijing, who are still friends and colleagues.
Later in 1998 I ended up working with a group of hard-core network systems engineers , based in Reston (Virginia, USA) as the Director of IT Products for Global One, and we created the first global IP VPN (Internet Protocol – virtual private network)backbone.
John – What made you decide to break from the large, corporate environment and strike out on your own?
Adil – I got tempted by the Internet boom. We had taken the Global One product team to its limit, and I wanted to participate on a more creative level as the Internet was really catching some good traction as global infrastructure. I went to work for a startup VoIP (voice over Internet Protocol) carrier, based out of Hong Kong as the head of product development. We did some amazing technical innovations, but got caught when the Internet bubble burst. That’s when I decided to go on my own and moved back to the UK to build my own network consultancy.
John – How important is innovation, first mover status, and taking technical risk? What would you advise IT managers or engineers to consider in their own companies?
Adil – I always use the term “… working right at the edge of the envelope…,” meaning the best place to be is absolute first mover status. As a startup, you must always consider maximizing new technology and innovation, with of course proper risk management. My advice to IT managers and lead engineers would be a “Calculated Offense” is your best defense. As a small company or startup company without innovation and managed risk, you put your existing services at company at a disadvantage. You must be able to discriminate yourself from the pack.
John – You’ve been with the Internet since the beginning, and lived each step of the evolution up till today. Are you comfortable with how the Internet has evolved? Mistakes made? Concerns with the current state of the ‘net? Happy with the Internet as it is today?
Adil – When I first got involved with rolling out the Internet into the Middle-East (NOTE: Adil was part of the telecom reconstruction team that went into Kuwait following the first Gulf War), Europe, and Asia it was exciting. I learned very quickly what a huge impact the Internet and Internet technology was going to have on people’s lives.
The evolution (of the Internet) was incredibly fast. And while I think along the way we could have done things more tactfully and strategically, my only regret is that in the early days the global carrier I worked for (Global One/Sprint International) didn’t fully commit to the Internet wave. We helped influence and change that later when I headed the product management, but global commercialization of the Internet had already taken off by the time my company fully engaged in building their network and product lines to meet customer and market expectations.
John – Where would you like to take the Internet, or more importantly, what does the world need from Internet and communications engineers to get where we should be in 15 years?
Adil – Back when I worked in Beijing and we built our first company Intranet using the IP protocol, we let everyone in the company go nuts with creativity and freedom of thought. It was a wonderful period, with a group of very talented and innovative Chinese engineers. We had not only the approval of our local and regional management, but also full support from the Chinese government which funded much of our lab work. I remember some other big corporations doing the same, and we called it the “chaos phase of the Internet.”
This is where we are again today with the Internet. In my opinion, Internet development now needs a bit more direction. In 15 years we shouldn’t have to work out how to plug into it (the Internet), it should be integrated and seamless anywhere. Connectivity and access to the global Internet should no longer be a burden, it should be a basic right of all persons in all countries.
Once we have cracked the nut of access, we will need to further force the IPv6 (Internet Protocol version 6) into the network, and better understand how to not only increase bandwidth in an affordable model, but also build in better efficiencies to eliminate bottlenecks. Only then can we really concentrate on encouraging open and creative development of applications that will bring our communities – regardless of geography and political controls – into the next generation of social and economic globalization.
John – You’ve always been a visionary, as well as an engineer. What do you think is the most important problem we have to solve with networks over the next 10~15 years, both technical and political?
Adil – Thanks for the complement! Technically our networks use similar components but work in very different ways, i.e. unique to the programming of hundreds and thousands of interconnected networks. The Internet was founded by establishing common ground rules, however we seem to have drifted away from some of the ground rules, and the processes in place to control the chaos of the Internet.
Those rules need to be re-established, but this is likely to be a political nightmare as governments struggle to gain greater control over both the Internet and people who are using the Internet. I do believe that from this period of chaos there will be a solution. The Internet itself is inherently self-healing, and from the chaos will emerge a stronger Internet.
John – What effect did your days in China, Kuwait, and other developing countries have on your desire to continue working in the developing economies of the world?
Adil – Working in developing countries I have the ability to leapfrog established thinking and technologies, and truly be involved in innovation. I can continue to be an engineer at heart and yet still drive technology, educate, create. All the things that help me maintain the “…edge of the envelope…” philosophy.
John – Where do you go from here?
Adil - Continue working with other visionaries and apply the results to real projects. I want to continue to contribute to the global community in any way possible. I is fun to actually see the results of your effort helping make people’s lives and futures more attainable. One of the marketing lines I used some years ago was “… I’m still working on a simple particle transportation platform…’ I think the future is going to be an exciting place. I have some ideas on what I might do next – still under wraps though…!
During his 20 year career, Adil Mehmood has served a wide range of senior roles in Global Telecomms with Tier-1 and Tier-2 telecommunications carriers, VoIP Carriers, Internet Service Providers and various specialist consultancies. He has worked in several international locations, and currently resides in Mongolia working for a Global Mining Company as their Enterprise IT Director.
Adil Mehmood holds a B.Eng (Hons) in Electronic Systems Engineering from Kingston University in the United Kingdom.
This post is for Nara – wherever you might be.
In early 2000 I visited some of my friends and industry colleagues in Ulaanbaatar, Mongolia whom I’d worked with for a few years on various Internet-related projects. In those days Mongolia only had 5 Internet service providers (ISPs), and 100% of Internet traffic was connected via low capacity satellite connections.
Each ISP had a separate connection, some to California, one to Hong Kong, one to Germany – there was really no planning other than to get the cheapest bandwidth possible to meet the needs of a rapidly growing Internet community. Mongolians use a Cyrillic-based written language, and Mongolians began to see the benefit of having a local economy that could function within the culture and language of the people.
The only problem was that each ISP had an independent upstream Internet connection, and for the most part the links were saturated. The interesting thing is that much of the saturation was due to Mongolian language or Mongolian interest traffic going out one ISP, to say an upstream connection in Stockton, Calfornia – and returning to another ISP’s user through an international connection based in Germany.
The result is as you would expect – very poor performance and user experience at a very high cost.
So hosting companies started to get smart. Rather than host web sites in Mongolia, companies began to find hosting platforms in North America and Europe to host Mongolian content. While not perfect, it did remove about one half of the performance bottleneck between users of different ISPs. Of course the bad thing is the revenue produced by hosting went to American and European companies, removed from the Mongolian economy forever.
So in early 2000 I met with friends and colleagues from several different ISPs in Ulaanbaatar, and brought up the idea of building a neutral Internet Exchange Point/IXP in Ulaanbaatar to facilitate local interconnection between ISPs. Intially there was a bit of reluctance to the idea of cooperating with competitors, but in the end the ISP owners relaized the customer performance and cost savings of taking local traffic off the international links made a tremendous amount of sense. A small Mongolian company called Infocon was chosen to manage the project.
Problem – no money to build the IXP. Solution – I used my credit card and bought a pile of Cisco switching and transmission hardware and donated it to the Mongolian ISP community. Another friend (Raphael Ho) came to Ulaanbaatar, configured, and connected the ISPs to what became known as the Mongolian Internet Exchange/MIX. The impact was immediate, and all of the reasons for building an IXP in a developing country met the model and image of how it should be done.
A few years later the original MIX was replaced by a high performance platform donated by another international group, and the MIX grew to support current robust Internet community thriving in Mongolia today.
The moral? The MIX represents a success story for remote locations and developing countries to use in ensuring their own economy and user community has the best possible tools available to reduce international transmission cost, increase end user and network performance and provide a positive experience. Why should a developing country pay a surcharge to the international community for developing a local economy? No reason at all.
For us who enjoy relative opulence in our worlds, consider the value we can bring to a developing country or company with our experience, and the way we can enable those developing countries to have a better chance to get up to global economy speed – often with very little effort of our own.
Some days it is fun to be an engineer.
The telecom world has once again dealt with the challenges of service disruption following cut or damaged submarine fiber optic cables. Over the past two weeks separate incidents in the Med caused by, as of yet undetermined causes, disrupted large volumes of both telecom and Internet traffic serving many countries around the Indian Ocean.
Most large telecom companies and carriers using large submarine capacity have redundant routes built into their networks. Thus if one of the major trunking links goes out of service, telecom traffic will immediately or dynamically reroute through another cable system. In some cases this is a much longer route, but service is usually not affected other than potentially increased latency on the circuits.
Internet traffic is a bit different. While Internet traffic may account for a very large percentage of global telecom capacity, the dynamic of Internet peering requires many more one-to-one relationships among other Internet network and service providers. Why? Because there are more Internet-enabled and Internet networks than traditional telecom carriers. An Internet network manager has to manage many more inter-company relationships than a traditional carrier network manager.
The need for redundant engineering within Internet networks is the same as any other telecommunications system. One tool the Internet community has available that is not shared by the standard carrier model are public and private Internet Exchange Points (IXPs). Wikipedia defines an IXP as:
An Internet exchange point (IX or IXP) is a physical infrastructure that allows different Internet service providers (ISPs) to exchange Internet traffic between their networks (autonomous systems) by means of mutual peering agreements, which allow traffic to be exchanged without cost. IXPs reduce the portion of an ISP’s traffic which must be delivered via their upstream transit providers, thereby reducing the Average Per-Bit Delivery Cost of their service. Furthermore, the increased number of paths learned through the IXP improves routing efficiency and fault-tolerance.An Internet exchange point (IX or IXP) is a physical infrastructure that allows different Internet service providers (ISPs) to exchange Internet traffic between their networks (autonomous systems) by means of mutual peering agreements, which allow traffic to be exchanged without cost. IXPs reduce the portion of an ISP’s traffic which must be delivered via their upstream transit providers, thereby reducing the Average Per-Bit Delivery Cost of their service. Furthermore, the increased number of paths learned through the IXP improves routing efficiency and fault-tolerance. (www.wikipedia.org)
Large IXPs such as the Amsterdam Internet Exchange and the London Internet Exchange have more than 300 members. All of those members have the option of peering with either individual networks, or large numbers of networks and Internet-enabled companies through use of route servers. While dedicated circuits between larger networks are still common (larger networks sell Internet transit to smaller networks, and thus will not normally “peer” at an IXP), peering at IXPs is very cost effective for networks requiring both direct relationships with other networks (eliminating the need for purchasing some transit traffic).
One of the most important statements in Wikipedia’s description deals with fault tolerance. As the Internet is a global network, and there is a need for global routing and access to every “end point” within the Internet, network administrators have a huge challenge designing effective disaster recovery plans for their networks. For example, if you are an Indian network provider, and your traffic is primarily routed through the Med and Europe, a cable cut in the Med can have catastrophic effect on your network performance and connectivity.
If the traffic going through Europe is further directed towards North America, then your network will have even more serious performance problem – if you are able to get any connectivity.
So, when the cable cut occurs, the network administrator may have to arrange restoral circuits with many individual network service providers, content delivery networks, VoIP companies, and other networks – if the Indian company does not have a physical presence at one or more large IXPs. At the large IXP, such as Amsterdam, London, Any2, or other large IXP, the network administrator has the option of quickly re-routing traffic through another cable directed at that large IXP, with the potential of many more peering relationships than if they were single threaded through a large trnsit carrier. This is particularly true of those carriers which may not have adequate restoral capacity or restoral planning for those circuits.
Our own IXP, the Any2 Exchange, normally sees large spikes of traffic during major cable disruptions. While there is a lot of primary traffic being sent through the IXP, it is also used by many carriers as a backup or tertiary disaster recovery point for Asian, Mid-East, Russian, and European networks to bypass disasters or cable disruptions. North America is unique in that both major Atlantic/Mediterranean and Asian cable systems connect to north America, allowing North America to be used as an alternate route for traffic transiting any continent.
While the American Internet community has lagged the Europeans in using IXPs (for a variety of reasons – to be continued in a later blog entry), both Asian and European networks are quickly increasing their presence at American IXPs. This is certainly paying off, as the European and Mid-East networks were able to quickly restore nearly all Interent traffic following our recent cable disruptions – much of that restoral going through the IXPs.
