Solutions by Workload

A second node doubles the memory but adds nothing to response speed, which stays tied to the 273 GB/s of each machine. We cover how two Sparks are stitched together, which QSFP112 cable NVIDIA validated, why the rated 200 Gb/s came out as 92 to 95 in open testing, and where the NVIDIA product page disagrees with the user guide about how many nodes can be linked.

A two or four node cluster is limited by the fabric, not by the cards inside the boxes. We go through the ConnectX-7 datasheet figures, the difference between an MQM9700 with an on-board subnet manager and an MQM9790 without one, what RoCE really takes on Spectrum switches, why a single OSFP cage carries two NDR ports, and the workloads where 100GbE closes the question entirely.

Not every team needs a rack and a server room. A workstation built on RTX PRO 6000 Blackwell 96 GB and Threadripper PRO 7975WX holds models up to 70 billion parameters, runs off a normal wall socket and sits in the office. We explain why 96 GB on a single card removes a whole layer of engineering, what ECC memory buys you, and why this is the natural step up from a DGX Spark.

When several models run on one machine around the clock, what matters is not peak throughput but the price per gigabyte of GPU memory. Four RTX PRO 6000 Blackwell Server Edition cards give 384 GB, and MIG hardware partitioning turns them into a dozen logical cards so teams stop queueing. We compare it with the H200 build and explain why a rack takes the server edition of the card.

Two H200 NVL cards of 141 GB each, joined by an NVLink bridge, give 282 GB of shared memory. We explain why 64 cores, why a terabyte of system RAM and four power supplies, how much memory a 70-billion-parameter model eats and what changes during fine-tuning. The configuration follows the standard set by Exxact, BIZON and Thinkmate.