Running the Cloud – statistic overload

A few years ago, when we first started the Live@edu email service, it was running on the same system as Hotmail. Since then we’ve moved it onto a completely Exchange-based system, which has actually been running Exchange 2010 for quite some time. And so I’d forgotten about Hotmail. But running cloud services at a massive scale requires quite a lot of work in the background, and I was surprised me when I read the “peek behind the scenes at Hotmail” article, on the Inside Windows Live blog, because there are some stunning stats about how Hotmail is now run:

  • We deliver localised versions of Hotmail to 59 regional markets, in 36 languages*
  • We host well over 1.3 billion inboxes.
  • Over 350 million people are actively using Hotmail on a monthly basis.
  • We handle over 3 billion messages a day and filter out over 1 billion spam messages.
  • We are growing storage at over 2 petabytes a month (a petabyte is ~1 million gigabytes or ~1,000 terabytes).
  • We currently have over 155 petabytes of storage deployed (70% of storage is taken up with attachments, typically photos).
  • We’re the largest SQL Server 2008 deployment in the world (we monitor and manage many thousands of SQL servers).

And the team go on to describe how they keep all of that running, and how they keep the deployment of new storage and systems ahead of the demand for it. I can’t imagine adding 2 million gigabytes of storage every month. That’s an awful lot of disks!

Gizmodo have a nice graphic which tries to put a petabyte into scale – 20 million four-drawer filing cabinets, or 13.3 years of HD-TV. And it equates 50 petabytes to the entire written works of mankind, from the beginning of recorded history, in all languages. And there’s three times as much as that in the Hotmail data centres!

* As I mentioned, our Live@edu service actually runs on a different system. For example, data for UK customers is stored in our new Dublin datacentre, rather than outside of Europe or simply spread over a range of different worldwide data centres

Comments (0)

Skip to main content