also the good thing about getting your own servers is having hardware options (like getting SSD drives instead of the regular ones, having a video card (if you do media transcoding), etc.)
But video transcoding can be faster by a factor of several, if done on a GPU. That's a big deal for those who do video transcoding all the time.
The irony is that GPU is a more powerful processor than CPU, just not as versatile and easy to use (only a small subset of programming tools can use it yet) but it changes.
Show me an open source encoder with GPU support. From my research, when you are doing high quality encodings the speed difference is negligible at best.
Also, at web scale encoding time isn't nearly as important as the ability to go parallel which is significantly cheaper and easier to scale using a bunch of EC2 instances.
We process 2-300 videos a month, though during competitions we process 10x that in a month. All of the bandwidth, storage, processing is less than $10K a year with an average encoding FPS of 33fps.
I would love pointers to sources if this isn't the case and I can do this for massify cheaper. Would be appreciated.
EDIT: We also encode 2-3 different versions of the video at varying sizes and bitrates.
There aren't any open source CUDA apps that do transcoding, so far they are all Windows apps. And non-scriptable ones at that, which rules out web scale.
My point isn't so much picking something off the shelf, its more develop something for yourself. If "web scale" means that the ROI is greater then development effort plus hardware acquisitions compared to your current system then do it.