Tesla has unveiled the latest version of the Dojo supercomputer, and it is so powerful that it tripped the power grid in Palo Alto. Dojo is Tesla’s custom supercomputer platform built from the ground up for AI machine learning and, more specifically, video training using the video data from its fleet of vehicles.
The automaker already has a giant NVIDIA GPU-based supercomputer that is one of the most powerful in the world. However, the new Dojo custom-built computer uses chips and an entire infrastructure designed by Tesla.
The custom-built supercomputer will elevate Tesla’s capacity to train neural nets using video data, which is critical to its computer vision technology powering its self-driving effort.
Last year, at Tesla’s AI Day, the company unveiled its Dojo supercomputer, but it was still ramping up its effort at the time. It only had its first chip and training tiles, and it was still working on building a full Dojo cabinet and cluster, or “Exapod.”
Tesla has unveiled the progress made with the Dojo program over the last year during its AI Day 2022 on September 30. The company confirmed that it managed to go from a chip and tile to a system tray and a full cabinet.
Tesla claimed it could replace six GPU boxes with a single Dojo tile, which the company claims costs less than one GPU box. There are six of those tiles per tray. Tesla said that a single tray is the equivalent of 3 to 4 fully-loaded supercomputer racks. The company is integrating its host interface directly on the system tray to create a sizeable full host assembly. Tesla can fit two of these system trays with host assembly into a single Dojo cabinet.