[dev] what if we succeed? « Tidepool News

Today I let myself dream a little, imagining what I would do if Tidepool attracted thousands of players.  Such “counting of chickens” has both practical and motivational value.  Planning for 25,000 users is daunting technical task in the short term, especially when you consider that popular Minecraft servers rarely (if ever) reach that limit.

I do have a few things in my favor. First, I’ve been designing for a large multiplayer environment from the start, unlike Minecraft.  Second, I’m using the same foundational tech that large companies like Bank of America use (the Java EE 6 stack).  Third, I’m an “application performance monitoring” (APM) consultant, having worked with companies like Marriott, Lowes, and (this week) Saks Fifth Avenue to help keep their LARGE systems running smoothly.

All things in my favor, but Tidepool is a different beast than these lean (and well funded) corporate systems.  It takes significant time and effort to test realistic load, especially for a system with graphics-intensive clients.  Firing up LoadRunner to simulate web users is one thing.  Imagining what’ll happen when 500 players decide to meet in-game for some community event is another.

Given this, I’ve been rethinking my “open the floodgates” date a bit.  Perhaps better to have a closed beta for a while.

| Main |

Leave a Reply

Your email address will not be published. Required fields are marked *