Scale and the ability to customize down to a granular level are the top factors behind the success of Amazon Web Services, according to the vice-president who builds data centres for the cloud provider.
Scale is the enabler of everything, said James Hamilton, vice-president of AWS, during the company’s re:Invent customer conference in Las Vegas last week.
Hamilton said when a company does some very often it gets very good at it. Following that logic, the more server AWS provisions the more it learns the best way to carry it out.
The same holds true for data centres, he said. By rapidly creating capacity, AWS is also allowing its engineers to test ideas faster.
Its work on storage and networking servers also enables AWS to customize to a very granular level. As a result, Hamilton said, AWS’s storage servers are far denser than anything on the market.
AWS even customizes its power consumption process. With more than 10 data centres in the eastern region of the United States alone, AWS is able to purchase power in bulk. It also has its own sub-stations that store power
This allows AWS to scale up as much as needed without worrying about running out of juice.