A deep know-how convention for processor and system architects from trade and academia has develop into a key discussion board for the trillion-dollar information middle computing market.
At Scorching Chips 2024 subsequent week, senior NVIDIA engineers will current the most recent developments powering the NVIDIA Blackwell platform, plus analysis on liquid cooling for information facilities and AI brokers for chip design.
They’ll share how:
- NVIDIA Blackwell brings collectively a number of chips, programs and NVIDIA CUDA software program to energy the subsequent technology of AI throughout use instances, industries and nations.
- NVIDIA GB200 NVL72 — a multi-node, liquid-cooled, rack-scale resolution that connects 72 Blackwell GPUs and 36 Grace CPUs — raises the bar for AI system design.
- NVLink interconnect know-how offers all-to-all GPU communication, enabling file excessive throughput and low-latency inference for generative AI.
- The NVIDIA Quasar Quantization System pushes the bounds of physics to speed up AI computing.
- NVIDIA researchers are constructing AI fashions that assist construct processors for AI.
An NVIDIA Blackwell speak, going down Monday, Aug. 26, may also highlight new architectural particulars and examples of generative AI fashions operating on Blackwell silicon.
It’s preceded by three tutorials on Sunday, Aug. 25, that can cowl how hybrid liquid-cooling options can assist information facilities transition to extra energy-efficient infrastructure and the way AI fashions, together with massive language mannequin (LLM)-powered brokers, can assist engineers design the subsequent technology of processors.
Collectively, these displays showcase the methods NVIDIA engineers are innovating throughout each space of information middle computing and design to ship unprecedented efficiency, effectivity and optimization.
Be Prepared for Blackwell
NVIDIA Blackwell is the final word full-stack computing problem. It includes a number of NVIDIA chips, together with the Blackwell GPU, Grace CPU, BlueField information processing unit, ConnectX community interface card, NVLink Swap, Spectrum Ethernet swap and Quantum InfiniBand swap.
Ajay Tirumala and Raymond Wong, administrators of structure at NVIDIA, will present a primary have a look at the platform and clarify how these applied sciences work collectively to ship a brand new normal for AI and accelerated computing efficiency whereas advancing power effectivity.
The multi-node NVIDIA GB200 NVL72 resolution is an ideal instance. LLM inference requires low-latency, high-throughput token technology. GB200 NVL72 acts as a unified system to ship as much as 30x sooner inference for LLM workloads, unlocking the power to run trillion-parameter fashions in actual time.
Tirumala and Wong may also talk about how the NVIDIA Quasar Quantization System — which brings collectively algorithmic improvements, NVIDIA software program libraries and instruments, and Blackwell’s second-generation Transformer Engine — helps excessive accuracy on low-precision fashions, highlighting examples utilizing LLMs and visible generative AI.
Retaining Information Facilities Cool
The standard hum of air-cooled information facilities might develop into a relic of the previous as researchers develop extra environment friendly and sustainable options that use hybrid cooling, a mixture of air and liquid cooling.
Liquid-cooling strategies transfer warmth away from programs extra effectively than air, making it simpler for computing programs to remain cool even whereas processing massive workloads. The gear for liquid cooling additionally takes up much less area and consumes much less energy than air-cooling programs, permitting information facilities so as to add extra server racks — and due to this fact extra compute energy — of their services.
Ali Heydari, director of information middle cooling and infrastructure at NVIDIA, will current a number of designs for hybrid-cooled information facilities.
Some designs retrofit present air-cooled information facilities with liquid-cooling models, providing a fast and straightforward resolution so as to add liquid-cooling capabilities to present racks. Different designs require the set up of piping for direct-to-chip liquid cooling utilizing cooling distribution models or by solely submerging servers in immersion cooling tanks. Though these choices demand a bigger upfront funding, they result in substantial financial savings in each power consumption and operational prices.
Heydari may also share his group’s work as a part of COOLERCHIPS, a U.S. Division of Power program to develop superior information middle cooling applied sciences. As a part of the challenge, the group is utilizing the NVIDIA Omniverse platform to create physics-informed digital twins that can assist them mannequin power consumption and cooling effectivity to optimize their information middle designs.
AI Brokers Chip In for Processor Design
Semiconductor design is a mammoth problem at microscopic scale. Engineers growing cutting-edge processors work to suit as a lot computing energy as they will onto a bit of silicon just a few inches throughout, testing the bounds of what’s bodily potential.
AI fashions are supporting their work by enhancing design high quality and productiveness, boosting the effectivity of handbook processes and automating some time-consuming duties. The fashions embrace prediction and optimization instruments to assist engineers quickly analyze and enhance designs, in addition to LLMs that may help engineers with answering questions, producing code, debugging design issues and extra.
Mark Ren, director of design automation analysis at NVIDIA, will present an summary of those fashions and their makes use of in a tutorial. In a second session, he’ll concentrate on agent-based AI programs for chip design.
AI brokers powered by LLMs might be directed to finish duties autonomously, unlocking broad purposes throughout industries. In microprocessor design, NVIDIA researchers are growing agent-based programs that may purpose and take motion utilizing custom-made circuit design instruments, work together with skilled designers, and study from a database of human and agent experiences.
NVIDIA specialists aren’t simply constructing this know-how — they’re utilizing it. Ren will share examples of how engineers can use AI brokers for timing report evaluation, cell cluster optimization processes and code technology. The cell cluster optimization work just lately received finest paper on the first IEEE Worldwide Workshop on LLM-Aided Design.
Register for Scorching Chips, going down Aug. 25-27, at Stanford College and on-line.