DigitalOcean And AMD Deliver Doubled Inference Performance For Character.ai

As enterprises seek alternatives to concentrated GPU markets, demonstrations of production-grade performance with diverse ...
Semiconductor Engineering

Efficient Multi-GPU Shared Memory via Automatic Optimization of Fine-Grained Transfers

Harini Muthukrishnan (U of Michigan); David Nellans, Daniel Lustig (NVIDIA); Jeffrey A. Fessler, Thomas Wenisch (U of Michigan). Abstract—”Despite continuing research into inter-GPU communication ...
dbta

Crusoe’s Atero Acquisition Drives GPU Optimization for Demanding AI Workloads

Crusoe, the industry’s first vertically integrated AI infrastructure provider, is announcing its acquisition of Atero, the company specializing in GPU management and memory optimization for AI ...

The Ultimate 3D Integration Would Cook Future GPUs

Peek inside the package of AMD’s or Nvidia’s most advanced AI products and you’ll find a familiar arrangement: The GPU is flanked on two sides by high-bandwidth memory (HBM), the most advanced memory ...
WinBuzzer

AI: Memory Bottleneck Emerges as Main LLM Inference Challenge

Google researchers have revealed that memory and interconnect are the primary bottlenecks for LLM inference, not compute power, as memory bandwidth lags 4.7x behind.
Electronics Weekly

Can memory-GPU stacking heat be beaten for AI training?

Belgian research lab Imec has revealed 3D stacked memory-on-GPU AI processor thermal data at IEDM (IEEE International Electron Devices Meeting) this week. The data comes from a thermal STCO ...
SiliconANGLE

Nvidia acquires GPU cluster optimization startup Run:ai for reported $700M

Nvidia Corp. today disclosed that it has acquired Run:ai, a startup with software for optimizing the performance of graphics card clusters. The terms of the deal were not disclosed. TechCrunch, citing ...