Tesla's New Supercomputer will Drive its Autonomous Push

[Harley Lover]

I'm hoping some of the techies among us will chime in on their capability, seems pretty impressive:

Quote

Karpathy claims Tesla's current supercomputer is the fifth most-powerful in the world. The new computer cluster's specifications include:

 

• 720 nodes of 8x A100 80GB (5760 GPUs total)
• 1.8 EFLOPS (720 nodes * 312 TFLOPS-FP16-A100 * 8 gpu/nodes)
• 10 petabytes of "hot tier" NVME storage @ 1.6 terabytes per second
• 640 terabytes per second of total switching capacity

Tesla uses the neural networks to label 4D data that comes from videos taken through eight onboard cameras that make up its vehicle's Vision system. That data is then used to train Tesla's software to autonomously navigate the car using only radar and the cameras.

However, Tesla recently said it would drop radar altogether, and began transitioning solely to the camera-based system in its Model 3 and Model Y vehicles in May. The ultimate goal of Tesla Vision is to make an autonomous car that is dramatically safer than the average person. 

 

Tesla might be alone in the auto world in abandoning radar/lidar - my impression is the OEMs around the globe seem to have agreed that LIDAR is essential for autonomous operation.

 

https://www.foxbusiness.com/technology/tesla-new-supercomputer-will-drive-autopilot

 

 

 

 

 

 

 

Edited June 22, 2021 by Harley Lover

[rperez817]

Video of Karpathy's presentation.

 

https://www.youtube.com/watch?v=eOL_rCK59ZI&t=29533s

[silvrsvt]

On 6/22/2021 at 8:54 AM, Harley Lover said:

I'm hoping some of the techies among us will chime in on their capability, seems pretty impressive:

 

What I can tell you hardware wise, is that its using the same architecture as a high end video card for the CPU