I'm hoping I can get some input on a little something I'm working on designing.
I'm in the (admittedly slow) process of designing and then building a scalable Athena deployment, and intend on making use of Amazon's AWS services. I've some experience with what they offer through other projects but haven't managed an Athena server of this scope.
I've attached a PNG output of a Visio diagram indicating how this would be designed.
It's designed to be secure, stable, and capable of growth based on user load (something that can be managed automatically by the cloud services provided once configured properly).
The key elements would be that each portion of the server would live in its own space, and all of it would be safe-guarded behind the cloud network's firewalls (with the web-faced being outside of course; web-site, forums, control panel), and that the map servers (from my understanding the part of the server that does the real work) would be contained behind a load balancer in sets.
The notation may be a bit confusing in the image but it's basically designed like this:
The game's world would be split over n number of map servers (I'm thinking 2-4ish) with each map server hosting a number of the game's zones. The load balancer would then monitor each of these individual zones, and in the even of being over-loaded automatically spawn a new instance of that particular server to assist in the original's load.
My question is... Will Athena make this unusably difficult to implement?
There's a major potential for cost savings if instances can be spun on and off based on need. Each of the [small] instanced servers are spec'd at roughly 2.3GHz/1.6GB.
Question
babillon
Hi everyone!
I'm hoping I can get some input on a little something I'm working on designing.
I'm in the (admittedly slow) process of designing and then building a scalable Athena deployment, and intend on making use of Amazon's AWS services. I've some experience with what they offer through other projects but haven't managed an Athena server of this scope.
I've attached a PNG output of a Visio diagram indicating how this would be designed.
It's designed to be secure, stable, and capable of growth based on user load (something that can be managed automatically by the cloud services provided once configured properly).
The key elements would be that each portion of the server would live in its own space, and all of it would be safe-guarded behind the cloud network's firewalls (with the web-faced being outside of course; web-site, forums, control panel), and that the map servers (from my understanding the part of the server that does the real work) would be contained behind a load balancer in sets.
The notation may be a bit confusing in the image but it's basically designed like this:
The game's world would be split over n number of map servers (I'm thinking 2-4ish) with each map server hosting a number of the game's zones. The load balancer would then monitor each of these individual zones, and in the even of being over-loaded automatically spawn a new instance of that particular server to assist in the original's load.
My question is... Will Athena make this unusably difficult to implement?
There's a major potential for cost savings if instances can be spun on and off based on need. Each of the [small] instanced servers are spec'd at roughly 2.3GHz/1.6GB.
Thoughts?
Link to comment
Share on other sites
3 answers to this question
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.