Nvidia Data Centers - The Data Processing Unit

2min Snip

00:00

Play full episode

Summary

Transcript

Episode notes

The melonocs chip is the third leg of the stool for en Nvidia. The company has beefed up its gpu and tensor cores to do machine learning specific, four by four by four matrix multiplications super fast. And they've put all this other non gaming, data center specific a i modules on to these chips and this hardware. Now what they're saying is, you've got your cpu, youv got your gpu. Now there's a d p u. That's like kind of born out of the melanox stuff, is the way that you really efficiently communicate and transform data within data centers.

By 2012, NVIDIA was on a decade-long road to nowhere. Or so most rational observers of the company thought. CEO Jensen Huang was plowing all the cash from the company’s gaming business into building a highly speculative platform with few clear use cases and no obviously large market opportunity. And then... a miracle happened. A miracle that led not only to Nvidia becoming the 8th largest market cap company in the world, but also nearly every internet and technology innovation that’s happened in the decade since. Machines learned how to learn. And they learned it... on Nvidia.

If you want more Acquired, you can follow our newly public LP Show feed here in the podcast player of your choice (including Spotify!).

Sponsors:
Statsig: https://bit.ly/acquiredstatsig24
Vanta: https://bit.ly/acquiredvanta
Crusoe: https://bit.ly/acquiredcrusoe

Links:

Ben Thompson’s great Stratechery interview with Jensen
Linus Tech Tips tests an Nvidia A100
Episode sources

Carve Outs:

The Expanse short story collection, Memory's Legion
Sony RX100 point-and-shoot camera

‍Note: Acquired hosts and guests may hold assets discussed in this episode. This podcast is not investment advice, and is intended for informational and entertainment purposes only. You should do your own research and make your own independent decisions when considering any financial transactions.