Scale Up, Scale Out Get a New Partner

By Bryon Moyer - 12 Mar, 2026 - Comments: 0

Key Takeaways: Three AI data center scaling strategies are scale-up, scale-out, and scale-across. Scale-up is within a rack; scale-out is between racks; scale-across is between data centers. Each of the three uses a different interconnect strategy to optimize either latency or jitter. As today’s data center workloads — especially for AI and HPC — outgrow the physical, ... » read more

Knowledge Centers
Entities, people and technologies explored

CPO Is Extending The Limits Of What’s Possible In AI Data Centers

Co-packaged optics technology will have a big impact on system power and the cost of data movement.

by Ann Mutschler

Flash Getting Stacked High-Bandwidth Version

Inspired by HBM, HBF could improve AI efficiency in 3D flash memory.

by Bryon Moyer

Can Edge AI Keep Up?

As models evolve faster than silicon cycles, experts weigh how much adaptability architects can afford without sacrificing power, area, or efficiency.

by Ann Mutschler

Chiplets Need A New Workflow

Multi-die assemblies are facing full system-level challenges, but engineering teams need coordinated and repeatable ways to identify risks early and scale reliably.

by Ann Mutschler

HBM4E Raises The Bar For AI Memory Bandwidth

The speed at which accelerators can be fed with data has become just as critical as raw compute capability.

by Nidish Kamath

AI Power on the Edge

Architecting solutions for edge AI is not about minimizing cloud solutions or making small extensions of existing MCUs/MPUs. It's a hardware/software/model co-development problem.

by Brian Bailey