An AI studio inference server requires careful matching of GPU compute, system memory, storage speed, and network t