Server Fault is a question and answer site for system and network administrators. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Our company currently has an openGL/directX powered app for viewing 3D models. Some of our customers have expressed an interest in a cloud based solution that would allow their users of devices with limited resources to view very large models which were being rendered on a server. The models we are looking to load would have a memory footprint of up to 2gb. In order to have decent performance we would like to use a hardware accelerated platform to render the images as the user manipulates the camera in some way.

I was looking at amazon web services as a potential solution but it seems that all the instances with GPU use the NVIDIA Tesla. Based on my reading this model of GPU seems more geared for crunching large data sets then real time rendering of 3D models. I have found some old posts from AWS developer forums in which people complain of performance issues with openGL rendering, which has led me to search for additional solutions, and ultimately here.

Has anyone had luck running an openGL driven app in the cloud on either AWS or another service? There seem to be streaming services geared towards games (such as NVIDIA Grid), which seems like a step in the right direction but which seem require substantial investment in hardware and infrastructure. We are hoping to find something with a simalr model to AWS where you pay for what you use.

Thanks for and help and suggestions you might have!

share|improve this question

closed as not constructive by mgorven, Ward, mdpc, Dave M, Khaled Feb 24 '13 at 7:44

As it currently stands, this question is not a good fit for our Q&A format. We expect answers to be supported by facts, references, or expertise, but this question will likely solicit debate, arguments, polling, or extended discussion. If you feel that this question can be improved and possibly reopened, visit the help center for guidance.If this question can be reworded to fit the rules in the help center, please edit the question.

This is such a special use case you'd likely be best off trying it out. – ceejayoz Feb 22 '13 at 20:40

I believe the reason amazon uses the GPU geared towards crunching numbers is because that's what you would use a server GPU for. There isn't a monitor hooked up to the server so the graphics portion of the GPU isn't all that relevant. When you remote into a server and/or stream, from a graphics point of view, to a large extent you are at the mercy of your client internet connection and the graphics capabilities of your client machine, not the server's.

That being said, if in your case the OpenGL rendering is the complex part, and if it can be done more efficiently remotely, and if what you actually want to see on the client can be sent over the wire from the server and displayed with your client graphics card in less time than it would take your client to render the data by itself, then it's possible you can benefit from this.

(Surely there must be a better way to write that last sentence...)

share|improve this answer

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