Distributed rendering lets you render an image or a scene using several networked computers or several processors on a multi-processor computer. Spreading out the workload this way considerably reduces the overall rendering time. Distributed rendering works best when you optimize the configuration for the strength and speed of your network and the power of each machine doing the rendering.
For more information about distributed rendering, see Distributed Rendering.
As a rule of thumb, several processors on fewer machines will yield better performance than several machines with few processors.
Each time you render a scene, start with a small number of machines and gradually increase the number until the extra machines provide no further benefit.
To improve distributed rendering performance, try using the machine with the most memory and the most powerful processor as the master.
Installing frequently used textures on all of the slave machines speeds up rendering by decreasing the amount of network traffic. This is especially true for memory–mapped textures.
Except where otherwise noted, this work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License
