Having increased our render capacity at one facility almost tenfold in the last five or six months, I feel like we’re close to the limit of what our setup can manage.
While this is probably one of those “how long is a piece of string” type questions, wondering what is considered the upper limit for how many machines can realistically be expected to run from a single repository, and whether we should be starting to thing about adding another setup in parallel?
What is your current node count at? That’s definitely a factor. The answer to your question really depends on your current infrastructure. We’ve seen 500 nodes rendering happily from one repository, but the server and the network need to be able to handle that load. We have a whitepaper that can help take the guesswork out of this sort of thing: thinkboxsoftware.com/deadlin … caling.pdf