Take the 2-minute tour ×
Game Development Stack Exchange is a question and answer site for professional and independent game developers. It's 100% free, no registration required.

When building a game server intended to accomodate large numbers of simultanous games (e.g. table card games), where the server manages lots of tables and each table has up to a small handful of users, what metrics are most useful to collect and why?

share|improve this question

closed as primarily opinion-based by Byte56 Jun 4 at 18:54

Many good questions generate some degree of opinion based on expert experience, but answers to this question will tend to be almost entirely based on opinions, rather than facts, references, or specific expertise.If this question can be reworded to fit the rules in the help center, please edit the question.

    
I think I need to clarify a bit. I am looking to things that people have found useful to track inside the game server. Things like number of active card tables, number of players, number of observers, avg. number of players per table, that sort of thing. I can keep going, but I am wondering what metrics you've found to be more useful than others and why. Thank you! –  Michael Teper May 26 '11 at 9:32

2 Answers 2

For metrics regarding the gameplay I'd suggest simply recording a full history of each game, with each action timestamped.

As card games tend to be fairly slow paced that shouldn't generate too much data and it lets you do any other analysis of the games you want to at a later date by just using that history data.

share|improve this answer
    
that makes sense and I already am saving the entire history. I was more interested in what would be useful to monitor the server in (near-) real time. –  Michael Teper May 23 '11 at 15:56

Since you're already tracking the state change of the game for each move, I don't see what else you might monitor in relation to the "game" itself; however, many commonly tracked server metrics include:

  • CPU Load
  • Physical Memory Use
  • DISK IOPS
    • Read Queue Length
    • Write Queue Length
    • Average Service Time
  • Network IO Use

Depending on your server platform there are different tools to monitor this data. The reason to monitor these, is because they directly impact performance and they are all finite resources on any given server. If you start maxing one of these out, you need do either, add more of that resource, re-write some of your application to work within the limits on your server.

share|improve this answer

Not the answer you're looking for? Browse other questions tagged or ask your own question.