Explosive growth of Internet and some questions

Explosive growth of Internet and some questions

Would building bigger data centers solve Internet’s performance problems?

Monday, September 01, 2008

Email This

Print This

Comments

RSS

If you are able to read this article without any glitches, thank Internet and its ever-improving delivery standards. However, performance problems continue to hurt Internet, which has to consistently face the ever-increasing traffic and perform better.

Michael M Afergan, CTO and Senior VP, Akamai Technologies spoke to Prasad Ramasubramanian of CyberMedia News on main issues of the explosive growth of Internet and what would drive this medium in the future. Excerpts from the interview:

The world works on platforms developed by Akamai and yet you remain reticent to the point that the hard work goes unnoticed. What do you have to say about it?

Thanks for recognizing us! At Akamai, we have always believed in the power of high value and highly focused marketing efforts. Our marketing strategy is B2B and we're not aiming to be a household name.

We believe the quality of service we provide and the value our customers gain are what's most important to our business. Our role lies in seamlessly overcoming the unseen challenges of the Internet middle mile and we are happy to let our customers to be in the forefront of these initiatives.

In most cases, Akamai's customers get noticed for the great work we do and that is exactly what we want. In most cases, we measure success against our customers' success and we are happy that this drives us as a company!

From developing gaming platforms in PS3 to addressing weaknesses in the Internet, what really has been the distinguishing point (read USP!) for your entity?

When you consider that Akamai today serves almost 20 per cent of the web traffic everyday and the rate at which Internet keeps evolving, you can appreciate the technology changes we need to keep making to be relevant in the market place.

In short, we want to provide the best possible performance, experience and availability for the customers of our customers and hence our strategy has always been to "deploy Web servers into thousands of ISPs close to end users".

Today we have deployed 34,000+ servers across 900+ networks. If you look at it, there are a couple of broad challenges, we as a company continue to keep addressing – how we distribute our network even more and how do we leverage the network in new and efficient ways. We are in 900 ISPs today and we would like to double that number soon.

We are not talking about peering; we are physically in those ISPs. If you are peering to get to the end users you are on the wrong side of the equation. You want to be where the end users are; you do not want to go through congestion points to get there. Secondly, we don't stop with having our servers near the end users. We sit on top of that network and find better routes on the Internet.

We don't change the protocol, but we will tunnel traffic through other Akamai servers that would not otherwise be on the path. This gives us the ability to find good routes on the Internet when the given route has congestion.

Finding a better route may mean we find multiple routes, so even if the route we are using experiences problems on the fly, we have another route running which will get the information where it needs to go.

Technologies like this and the intelligence they throw up have made it possible to deliver both applications and content and this has become a huge differentiator for us in the market place. At a very high level, we are not an ISP, so we do not route traffic in the traditional sense.

How we move packets from one server to another and how we direct end users to get to a machine that is very close to them, how we handle load balancing and fail over and what functionality we get into servers near end users… these are the key problems we solve. Originally it was static content but now dynamic content and full applications can run on Akamai's distributed network. That is what makes us unique.

What would drive the growth of Akamai in an eon where Internet has matured in most of the markets across the world and online commerce and video online content ready to take the Internet journey further?

You talk about how the Internet has matured. The Internet has grown in capacity but in disproportionate ways. The content delivery capacity has gone up by a factor of 50 to end-users. I had a 256K modem, now we have 20-40 Meg service. On average, over the last five years, you have had a factor of 50X growth in the last mile of bandwidth.

The first mile bandwidth, where the data centers are located at the core of the Internet, has grown by a factor of 20. You think about five years back, a typical customer might have a T3, 45 MB for the bandwidth. Today that same customer has GiG-E connectivity. More or less a factor of 20 there and it is not keeping up with the last mile.

The big problem with the Internet has been the middle mile. The peering points between the big networks and the smaller local networks, the big hauls, are the main problem area. That area has, more or less grown by a factor of 6.

The reason is that money is flowing into the Internet in the first mile, because enterprises pay for the bandwidth they use. Money flows in at the last mile because end customers pay for that.

The middle mile clearly does not have a business model supporting it to evolve and this is the problem Akamai solves! Video increases our traffic which is good for our business. It increases the burden on the Internet, which is bad for the Internet.

As we think about growing our capacity over the next few years we were thinking on the order of a couple hundred terabytes per second to support a large audience for TV or better quality of video. The only way you can do that is to be serving that content locally and that is where we come into the picture. You are never going to get that capacity out of data centers.

E-commerce is a great new source of revenue and efficiency, but if we don't solve some of these problems, some of that may be at risk. The first problem is the Internet really doesn't have much security integrated. The basic protocols are insecure.

The bad guys have figured out that there is a security loophole and they are exploiting it to conduct wide-ranging forms of attack -- everything from sending you to a spam site to identity theft to phishing and pharming.

That said, Akamai can help shelter our customers from a lot of these issues. We can help shield them against DoS attacks. We can help make their site more reliable. We think of it as how to provide a real physical shield around the infrastructure.

Akamai directs the end user request to our network that has over 34,000 servers. So there isn't a central point of failure. There isn't a central critical point to attack. Secondly, e-commerce sites today are becoming more personalized and dynamic, and users cannot wait for more than a few seconds to view content and run applications.

We help these retailers, having rich and dynamic websites through our Dynamic Site Solutions, enabling them to deliver a high-quality user experience, increasing browser to buyer conversions, generating additional revenue, and enhancing the overall effectiveness of their websites.

From ensuring that the right advertising content is served on Web pages based on geography, to managing TV blackout rules for online distribution of sports content, Akamai does everything. Who are you competing with in this space and could you tell your market share or web space?

We have always existed in a very competitive environment. When we have a prospect it is not that they do not have a website. They have always something sophisticated up and running and the first question is why should I outsource it at all?

Build versus buy is the number one source of competition for us in the market, primarily because some companies try to scale and manage their traffic by building infrastructure on their own.

The second level of competition is the managed service provider where the pitch is basically I can do a little of everything, and maybe that should be good enough.

And the third is a direct competitor who says, look I am a specialist too. We do very well competing against all three categories.

Some of those competitors are public, some are private, and some are very larger companies. We focus on what our customers' need, on meeting their needs to make their business models work and a few core segments and we love competition.

You have studied the impact of economic incentives on networked systems, reliable networking and content distribution protocols, and wireless protocols and architectures. What is your study on the slowdown that the globe is witnessing and how would it affect the functioning of Internet arena?

These are tough economic times no doubt, but I don't think Internet and businesses supporting the Internet like us will be hurt by this. We expect companies to save money by moving operations to the Internet and businesses will see moving to the Internet as a good strategy to cut costs in a tough economic climate.

©CyberMedia News

Email This

Print This

Comments

RSS

C-Change 2009

Meet the top Decision makers of Large enterprises from the 18th to 20th Feb '09

Explore the partnership options now

Stretch the dollar!

When budgets are slashed but the targets are not, you need greater ROI. Take a look at this special advertising offer from CIOL.

know more

	Make a life with software lifecycle
	Try Airtel Conferencing now!
	Strong Business needs Green Intelligence

	Motorola : Good Mobile Messaging
	Smart Enterprises = Smart CIOs
	"Start, Build & Advance your PM career"

	NSA: Network security & performance
	How to fix web application security and compliance issues

	Integrate your diversified system environment - Wipro
	HP Business Availability Center
	Transforming Business