InfiniBand

5 statements across 4 episodes · 0 bullish · 2 bearish · 4 people on the record · first statement Feb 19, 2024 by Erik Bernhardsson · said 12 times in 5 episodes since 2024 · across every show →

Mentions by year

brought up most by Jonathan Frankle (4), Evan Conrad (3), Kyle Kranen (2), Soumith Chintala (1), Akshat Bubna (1)

tap a year for its mentions
003162202420252026episodesmentions
012202420252026episodes it came up in
001.5132202420252026episodesmentions per episode
2026 3 mentions in 2 episodes 2 per episode
2025 3 mentions in 1 episode
2024 6 mentions in 2 episodes 3 per episode

every mention, scene by scene, with the transcript →

Everything said about InfiniBand, oldest first

Feb 19, 2024 negative
Disclosure
Modal Is Not Designed for Training Large Foundation Models
“When you get to like training, like very large foundational models, that that's a use case we don't support super well because that's very high IO, you know, you need to have like infinite band and all these things. And those are things we haven't supported ye…”
Erik Bernhardsson Feb 19, 2024 ▶ 27:44 Truly Serverless Infra for AI Engineers - with Erik Bernhardsson of Modal
Jun 25, 2024
Assertion Not checkable as stated
Frankle: Porch pirates stole MosaicML InfiniBand cables twice before data center delivery
“Our InfiniBand cables getting stolen from the data center twice, like in boxes before they arrived at the data center, like, you know, porch pirate basically had stolen our InfiniBand cables back when those were hard to come by”
Jonathan Frankle Jun 25, 2024 ▶ 10:54 State of the Art: Training 70B LLMs on 10,000 H100 clusters
Jun 25, 2024
Insight
Frankle: Large-scale model training forces teams to debug the full infrastructure stack
“It's kind of impossible if you're doing training to not go all the way through the entire stack, regardless of what happens. Like somehow I'm still chatting with cloud providers about power contracts, even though the whole point of dealing with the cloud provi…”
Jonathan Frankle Jun 25, 2024 ▶ 23:07 State of the Art: Training 70B LLMs on 10,000 H100 clusters
Apr 11, 2025 bearish
Prediction Not checkable as stated
Conrad: Decentralized compute networks will never beat co-located InfiniBand clusters
“I just don't really think this is gonna ever be more efficient than a fully interconnected cluster with Infiniband, or, you know, whatever sort of next spec might be. Like, I could be completely wrong, but Speedolite is really hard to beat. And regardless of w…”
Evan Conrad Apr 11, 2025 ▶ 34:33 SF Compute: Commoditizing Compute
Mar 8, 2026 neutral
Assertion Supported
Scaling beyond eight NVIDIA H100 GPUs requires inter-GPU communication over InfiniBand
“The maximum NVLink domain, domain for H 100 for most DGX H 100 Is eight GPUs, right? So if you scaled up past that, You're gonna have to figure out ways to handle the fact that now for the GPUs to communicate, you have to do it over InfiniBand, which is still …”
Kyle Kranen Mar 8, 2026 ▶ 30:26 Agent Inference at the "Speed of Light" — How NVIDIA moves like a $4.3 Trillion Startup
Made with StarZero

Turn any episode into a week of clips.

This entire site, over 200 episodes transcribed, diarized, checked and made playable, runs on the StarZero media pipeline. Drop in your own episode and the podcast clipper finds the moments worth sharing, cuts them, captions them, and reframes them for every feed.