Menu
Why the cloud requires a totally different data center

Why the cloud requires a totally different data center

If there's one mind-blowing statistic about Amazon Web Services, it's the company's scale.

If there's one mind-blowing statistic about Amazon Web Services, it's the company's scale.

The cloud is a nascent technology, but AWS is already a multi-billion-dollar business and its cloud is reportedly five times bigger than its 14 top competitors combined, according to Gartner. Amazon's Simple Storage Service (S3) stores more than a trillion files and processes 1.5 million requests per second. DynamoDB, the AWS-designed NoSQL database, is less than a year old and last month it already had more than 2 trillion input or output requests.

Supplying all those services at that scale requires a lot of hardware. The cloud division is growing fast though, which means that AWS is continually adding more hardware to its data centers.

How does AWS keep up with all that? The man who directs the strategy behind it, AWS Vice President and Distinguished Engineer James Hamilton, shared insights into this at the company's re:Invent customer conference in Las Vegas last week. In a nutshell, "Scale is the enabler of everything," he says.

[MORE AWS:Amazon bashes private clouds, launches a virtual desktops]

Scale helps the company in a number of ways, Hamilton says. For one, when a company does something a lot, it gets really good at it. So, the more servers AWS provisions, the more it learns about best practices in doing so. The same is true for data centers. The company is adding so much capacity so fast that it allows AWS engineers to test ideas quickly to find out what works and what doesn't, Hamilton says.

AWS has optimized its hardware for its specific use cases, he says. AWS has built custom compute, storage and networking servers, which allow the company to customize down to a granular level. Its storage servers are "far denser" than anything on the market and each weighs more than a ton, Hamilton says. Most recently AWS customized its networking gear to create routers and protocol stacks that provision high performance workloads. Hamilton says those investments will drive new networking-based products and services that will be released in the coming year. The idea of creating customized hardware is not unique in the industry - Facebook has created an entire open source project named the Open Compute Project for this process.

AWS even customizes its power consumption processes. The company has negotiated bulk power purchase agreements with suppliers to get the energy needed to power its dozens of data centers across nine regions of the globe. The US-East region, the company's largest, has more than 10 data centers alone, Hamilton says. Not only does AWS buy power in bulk for these data centers, the company has its own sub-stations that store power before it's served to data centers. This allows the company to be able to scale up its offerings as much as it needs to without worrying about power being a limiting factor.

AWS says a big goal is to be responsive to customers. The company recently launched a new US-West data center in Oregon that is 100% carbon neutral and Hamilton says it's the company's fastest growing data center.

Even with all the customization, AWS can't always predict exactly how much of its resources will be used. If AWS can increase its utilization, its costs will be lower because it will get more bang for its buck from the hardware. AWS has to have enough capacity to support its peak utilization, but when the servers are not being maxed out, that is a waste of capacity. AWS solves this problem by pooling dissimilar customer workloads. For example, retail customers may have one really busy time of the year during the holiday shopping rush, while companies handling tax returns use resources at another. By combining the workloads of these customers under the same umbrella, the workload density averages out.

There will still be under-utilization, but AWS has tried to turn that into an advantage. The introduction of spot-instances, which allow customers to place bids on excess instances, enables this.

It's these efficiencies that are gleaned from the company's scale that have fueled growth. As it grows, AWS adds more features, which attracts more customers, which leads to more growth. It's a virtuous cycle that AWS executives say drives the company.

Senior Writer Brandon Butler covers cloud computing for Network World and NetworkWorld.com. He can be reached at BButler@nww.com and found on Twitter at @BButlerNWW. Read his Cloud Chronicles here.  


Follow Us

Join the newsletter!

Error: Please check your email address.

Tags Amazon Web ServicesConfiguration / maintenanceGartnerCloudhardware systemsData Centercloud computinginternet

Featured

Slideshows

HP channel recognised at 2017 Partner Awards

HP channel recognised at 2017 Partner Awards

The HP Partner Awards 2017 at Shed 10 kicked off with an AMD-sponsored hackers lounge, a mysterious gaming style area filled with dry ice and red lasers, the waiters wearing Mr Robot style masks.

HP channel recognised at 2017 Partner Awards
Tech industry comes together as Lexel celebrates turning 30

Tech industry comes together as Lexel celebrates turning 30

Leading figures within the technology industry across New Zealand came together to celebrate 30 years of success for Lexel Systems, at a milestone birthday occasion at St Matthews in the City.​

Tech industry comes together as Lexel celebrates turning 30
HP re-imagines education through Auckland event launch

HP re-imagines education through Auckland event launch

HP New Zealand held an inaugural Evolve Education event at Aotea Centre in Auckland, welcoming over 70 principals, teachers and education experts to explore ways of shaping and enhancing learning using technology.

HP re-imagines education through Auckland event launch
Show Comments