Hedgehog gives AI cloud builders fast job completion time, low latency inference, and a convenient cloud UX with a turnkey software appliance for an AI network.
Mark your calendars! @githedgehog is taking the stage at Networking Field Day 35 on July 10-1 to showcase the latest innovations in enterprise IT networking. Don't miss the opportunity to learn how Hedgehog's cutting-edge solutions are transforming the industry. Learn more: https://t.co/YoyO7VOgIf
@engineering Well done. Love your cost optimization. @githedgehog can help you further reduce cost with open networking for your on-prem GPU supercompute clusters.
OK Seattle - #FoundersBash 2023 registration is open!
Last year we had 400 founders in the bldg to chill w @djtopspin vibes, good food and zero content or salespeople :-)
Remember, FOUNDERS ONLY. We're checking.
(VC friends DM for access)
https://t.co/5Iu8JV1mhu
3. There is going to be a massive migration of GenAI/LLM activity from 3rd party models (like OpenAI) to on-prem proprietary models for nearly all enterprise use cases.
[5/7] Fabric! A thin blue cable is optical/AOC (expensive) and a thick black one is copper/DAC (cheap). DAC only works short range ~5m.
The data NICs flow into an underpopulated Arista 6070CX switch, and out-of-rack with 8x AOCs.
gpu-gpu is all AOC, maybe to a core switch.
@timzaman Green if you don't care about ever catching a salmon again on that river. Was hoping to see immersion cooling here at least for lower power consumption.
Azure/Microsoft released pics of OpenAI's next-gen AI datacenter.
This supposedly powers [Chat]GPT* training (inference?). From the pics we can infer most components, power, network/ib topology, layouts, vendors and more.
A thread 🧵[1/7]
@timzaman This switch should be running SONiC like the rest of Azure. @githedgehog provides an open network fabric on SONiC if you want to run a similar fabric for your own AI workload.