Amazon counts on ‘courage and innovation’ to face the rise of AI

The AI ​​revolution is upon us, and companies around the world are looking to dive headfirst into the technology made famous by ChatGPT.

To meet their need for generative AI, companies must equip themselves with cutting-edge software and increase their computing power, which quickly leads them to knock on the doors of global cloud computing giants, of which Amazon’s AWS is the largest.

Amazon’s computing arsenal is housed in data centers around the world, and Prasad Kalyanaraman, vice president of AWS infrastructure, is the man in charge of keeping them running.

Amazon’s AWS data centers are spread across dozens of regions, serving as a sort of engine room for the online world, with Microsoft and Google being the company’s closest rivals.

And with the generative AI revolution entering hyperdrive, it’s up to Kalyanaraman to ensure the data center’s battalions are ready for the challenge.

“It takes a lot of courage and innovation” to meet today’s computing needs, Kalyanaraman told AFP in an interview at Amazon’s second headquarters near Washington.

“Building the right technology, both in terms of consuming the least amount of power needed and optimizing the entire process, from the chip level all the way to the data center level… requires a lot of innovation,” he said.

Kalyanaraman, a graduate of the prestigious Indian Institute of Technology and Queen’s University in Canada, has been with Amazon for nearly two decades, working on software before holding the keys to the data centres.

“Most people today are using cloud computing without even knowing it. If you visit a website, stream a video or check your transactions at your financial institution, you are actually using some form of cloud computing,” he said.

Amazon’s decision to make cloud computing a sideline dates back to 2006, when the company realized that its partners and vendors didn’t want to build — or buy — expensive computer networks.

The story continues

“We found that it’s very difficult for our customers to… tackle all the challenges of building that infrastructure. So why not bring that to them,” just like utilities bring electricity to your home, he said.

Nearly two decades later, AWS accounts for nearly 20% of the giant’s total revenue and generates about two-thirds of total profit.

– Constraint as opportunity –

“It’s a pretty big undertaking to build a data center from scratch,” Kalyanaraman said.

“First of all, you obviously have to find enough land to be able to deploy these data centers. Typically, we deploy them further away from metropolitan areas,” for both cost and environmental reasons, he said.

Connectivity is also key, as most customers want high computing speeds while being closer to their data.

Then you need a power source and power lines to get the electricity.

With success comes scrutiny or, in the case of some communities around the world, exasperation with the proliferation of data centers.

Data centers can intrude into a region’s bucolic landscape and place a significant strain on local power supplies, straining already fragile electrical grids.

And with the emergence of generative AI, Amazon has announced new projects around the world.

Kalyanaraman acknowledged that “energy is going to be a limited resource in today’s world, especially with generative AI and some of the other things that are required to run that amount of computing.”

But while “it’s not something you can really change overnight,” Kalyanaraman said AWS has worked with power companies to manage the flow, including through renewables.

AWS “is now the largest purchaser of renewable energy in the world for the fourth year in a row,” he said, as AWS has committed to becoming a net-zero carbon company by 2040.

Ever the technology optimist, Kalyanaraman remains confident that innovation could find a way to meet the challenge of generative AI, with the industry looking to nuclear power for help.

“Every time we have been faced with a constraint, we have all found a way to innovate.”

“I see (AI) as an opportunity,” he said.

arp/md