NBN just one piece in broadband puzzle: Akamai

Providing 100 megabit-per-second (Mbps) fibre connections to 93 per cent of the Australian population on the National Broadband Network (NBN) will not guarantee superfast speeds, unless the rest of the network is kept up to scratch, according to content delivery company Akamai's co-founder and chief scientist Dr Tom Leighton.

Tom Leighton
(Credit: Akamai)

The man who co-founded the content streaming platform company in 1998 during his time as a professor of applied mathematics at MIT in Boston, Massachusetts, said the Australian Government's AU$35.9 billion project to roll out fibre to the home was a good initiative, but said that people needed to look beyond just the last mile roll-out.

"I think it very good, because it gives connectivity across the last mile to a lot of people and it enables new industries, [but] the challenge with that, is that a lot of people think they just get their 100Mbps connection across the last mile, and they're done."

"If you're not careful, performance can degrade because so many people now expect to get high bandwidth and think they can use it. That will place an incredible demand on the peering and on the core infrastructure," he told ZDNet.

"With the traditional models of delivery, putting 100Mbps links into all the homes doesn't solve the whole problem. In fact, it puts more pressure on the infrastructure, and that becomes the bottleneck."

He said this was the reason Akamai was created, and with 95,000 servers worldwide, 150 points of presence in Australia alone. The company's goal is to make content available to users locally and efficiently.

"We want to be across the last mile of every end user on the planet. Wherever there is a cluster of end users, we want to be in their city, in their network and close to them in their city and network."

Using the vast amount of content that travels over its platform, — Leighton estimated that between a third and a sixth of all internet traffic travels over Akamai's platform — the company is able to work out more efficient ways to get content across the globe.

"We deliver over 20 million items per second. In a typical day, we'll have between 2.5 and 3 trillion deliveries. Every delivery is logged and delivered to the customer," he said.

"We use [the data] to better route traffic on the internet, so it makes the internet be more efficient. We use it to make deliveries faster; we use it to figure out where to place our servers."

Starting up

Telstra CEO David Thodey remarked this week that, to get good innovation and start-ups in Australia, the country just needed more self-belief in what it was doing.

"I think it is self-belief. I think it's about creating an environment where [innovation] is seen as good, we're encouraging it, creating a vision for it," he said. "And success helps; if you can get a few good wins under your belt, then the industry will grow in its own right."

Leighton said that working around people who had success before definitely helped get Akamai up and running, from operating in academia to being a company in 1998.

"Part of it is the chicken and the egg. When there are a lot of people around you that have done it, and there's news about it in the environment, it becomes an acceptable thing to do; you can talk to people who have done it," he said.

"Part of it is critical mass. It takes some effort to actually create all that, but once you do, you're in a very good state because the engine is there and it can crank out a non-trivial number of success stories."

Akamai's creation was thanks to the US$50K Business Plan competition at MIT. Leighton said that this year-long competition educated the company's founders in how to write a business plan, and put them in touch with potential investors and customers.

"We really didn't have any intent on forming a company. We liked the technology and we were hoping people would use it in industry, but we didn't think we'd ever form a company. The year-long event really provided us a stepping stone and the know-how to create a company, which we ended up doing in August 1998."

"It really is, for us, the bridge between an academic idea, resulting in academic papers and the creation of the company."