Twitter Needs Amazon AWS

It’s no secret that Twitter has its hosting flaws. It was all over the ‘net this morning that the Apple Keynote Killed Twitter. Since Twitter doesn’t seem to announce its outages well, I’ll help them out a little here and tell you why they went down.

I follow the Twitter Developers Group on Google, and one of Twitter’s head developers (I believe), Alex Payne, explained the situation:

“We seem to be up at the moment, but we had a flood of traffic around
the MacWorld Keynote this morning. No surprise, of course, but since
we couldn’t move data centers earlier this month we don’t have many
machines to throw at the increased load.

In the meantime, I’ve drastically increased the API rate limit and
disallowed unauthenticated requests to the API. This is temporary,
and once our load decreases later today things will be back to where
they used to be.

Thanks for your patience, and enjoy the Apple fanfest! “

“but since
we couldn’t move data centers earlier this month we don’t have many
machines to throw at the increased load.” Alex – you guys really need to consider having a few Amazon AMIs on hand in cases like these! It would have taken you about 5 minutes to have a copy of your current servers up on Amazon in the flood of Apple traffic were that the case. Twitter just got a round of investment a few months ago. This would be a great (and cheap!) use of that investment!

It appears Twitter is back up to normal, although it seems a few things are still trying to catch up.