How To Deliver Computing at Scale: Is this the Future of Data Centers?
Intel recently released a report that, if correct, has interesting implications for the future of data centers and the enterprises that use them.
We’ve seen the rise of “Big Data.” We’ve seen the way that virtualization became de rigueur and far more time is spent tweaking resource allocation than ever before. The inner workings of companies’ IT infrastructure are ever more massive, ever more harmoniously orchestrated. There are fewer and fewer silos, and for good reason — it’s a poor use of resources.
Intel sees this as a paradigm shift for companies that has bedrock implications for the data centers that serve them.
According to Jason Waxman, Intel’s General Manager of the Cloud Platforms group, “We see that the world is moving to scale computing in data centers. Our projection is between 70 and 80 percent of the compute, network, and storage will be going into what we call scale data centers by 2025.”
What are Scale Data Centers?
“Scale” data centers aren’t exactly new. They’re a data center configuration that’s used primarily by web giants like Facebook, Google, and Microsoft. The design allows these companies to deliver computing “at scale,” which is to say that they’ve been completely designed and optimized for this task. It is the ultimate priority.
It’s easy to see why, when you look at where the world is heading: We’ve got cloud-based applications in absolutely every facet of life. We’ve got self-driving cars that need to talk to each other — not to mention the machine learning that takes place. Personalized medicine (and all Big Data applications) requires large-scale deployment as well.
Basically, of all the reasons to lease space in a data center, small-scale deployments are a smaller and smaller size of the pie. Already, Intel estimates that 40% of the world’s IT capacity is currently located in scale data centers. When you think about it that way, 70-80% doesn’t seem like much of a stretch, especially when you take a step back and look at how demand for computing power has grown in the last decade.
The Opportunity in Scale Data Centers
The benefits of information don’t come from owning it, or even from accessing it, so it makes sense that the overriding priority would be the ability to parse the information. Crunch the numbers. Analyse the data. Give your machine learning programs an opportunity to mow throw it like a baleen whale through a school of krill.
Whatever you want to call data centers efficiently optimized for that kind of performance, that’s where we’re headed.