We’d like to run one last test. The hosting provider is on board to monitor performance, and we’re standing by to analyze the data. At 8pm EST 12/2, we’d like as many people as possible to hit human-registration and we’ll see what happens. While we know the load won’t be equivalent to the load during a real sale, we believe it will be a reasonable model we can extrapolate from. The hosting provider has more tools looking at our system so we expect to be able to dig through the results and make a better informed decision.
Just a note on load. We’ve been going back through previous years logs and comparing hit rates. In the past our older site was a bit simpler, but even when we take that in to account, the turn out this year to buy tickets is huge. We’re seeing the equivalent of 3x the traffic we’ve saw last year in round two. That’s a pretty serious increase. And again, as point of reference, this is at least 50x greater traffic than we had in 2010 for the 2011 sales process.
We appreciate your help in our crowdsourced load test. It’s been an interesting few days and we’ve had some really good discussion with folks about our process, the tech behind it, and the con overall. We’ll be posting a something longer on the overall philosophy on ticket sales later tonight.