Championship Servers

Did I miss the poll about the improvements on the servers?

If we choose Admins, Tick Speed, Tick Limit… Think we also should choose better servers?


Alrightt dude, which servers do you want? you wish to move on AWS? Azure?

1 Like

Kind of an ironic post but:

  • We’ve moved to Google Cloud, which sports the best latency of the cloud providers and great protection from DDoS attacks along with good performance, back in 2019.
  • BD over the years gained more and more data, which ultimately lead to us having 6+ downtimes a day. We managed to dip beneath 95% uptime. On the good months. Ouch.
  • We’ve roughly 99-100% uptime depending on the month you check. Sometimes we have an issue or some part of our line ends up clogged, but even in the worst weeks, we’ve not managed to get under 98% for that week except, as far as I recall, that time Google itself was down.

Even in our bad months, our uptime is pretty stellar now. Our latency is half what it used to be all around the world, ditto for our response time. We’ve solid backup policies and we’ve neither the random tick breaks in early BDNC, the DDoS issues or crash issues from mid-BD nor the frequent overloads of late-BD.

It’s not perfect, but my point is, we’ve made great strides. Especially when you consider we’re working with a 10 year old game that was meant to be replaced within a year when it was made. It’s filled with superfluous data, which we tried to clean but just cleaning a bit of it took almost a full day of downtime… We’ll still look into it again, once we feel that’s the only bottleneck.

But the truth is the game is just old, it was a pain to get it to work as well on a new server as on an old one (servers from 10 years ago were faster at single-threaded tasks, the type BD runs off, than modern ones), and that we managed to improve the tick engine’s calculating power by roughly 20-30% is something I consider an achievement.

So, tl;dr, its not perfect, we want it better, but let’s not forget how bad it used to be. We’ll get there, we’ve handled everything else thrown at us…

Now, as for CE, we’ve made huge changes to E1 (the CE world) as to further improve its performance. The tick engine for it should be blazingly fast (go try it - it’s 12 t / hr for now!) and rather stable as well. We’ve also alleviated the issue we had today with performance already as well, around an hour and a half ago and about an hour or two after it was identified, which isn’t perfect, but isn’t terrible either.

We’ll continue to take steps as we see a need for them, and continue to monitor the situation. We’ve a few contingency plans incase CE ends up handling worse than expected.

Hope that explains. I