AI Infrastructure
AI infrastructure and compute ecosystem
Articles
EBook – Accelerate Silicon Design for Physical AI (Part 1)⭐9
Discover how silicon design is evolving for Physical AI in 2026—real-time processing, energy efficiency, and safety innovations.
Chip Industry Technical Paper Roundup: Sep. 1⭐9
Chip industry technical paper roundup: latest research, innovations, and breakthroughs in semiconductor engineering from September 1.
Workload-Driven HBF Substrate for Capacity-Scalable LLM Inference⭐8
Workload-driven HBF substrate enables capacity-scalable LLM inference with 3.2x bandwidth gains and 40% lower energy.
Trust, But Verify: Ensuring Reliability in AI-Driven⭐9
Discover how AI transforms chip design and verification in 2026, and why robust strategies like formal and cloud-based testing ensure reliability.
Cycle-Level Simulator for Distributed GPUs in AI Workloads (Purdue)⭐9
Cycle-level distributed GPU simulator for AI workloads, modeling compute, memory, interconnects, and power for 2026 data center design.
M3D 6T SRAM with BEOL Pass-Gates at 2nm: A Collaborative⭐8
Georgia Tech and Synopsys unveil 2nm M3D 6T SRAM using BEOL pass-gates, boosting cache density and energy efficiency.
Hybrid HBM-HBF Architecture in LLM Inference⭐9
Hybrid HBM-HBF architecture for LLM inference cuts memory bottlenecks. Oxford research combines high-bandwidth memory and flash for scalable AI performance.
Predicting Post-Route PPA from Macro and Standard-Cell Placements⭐9
A methodology predicts post-route PPA from macro and standard-cell placements, cutting routing iterations and enabling faster, informed physical design decision...
Rowhammer Backdoor Injection Attack During Inference: A New⭐8
A new attack from Northeastern University exploits Rowhammer to inject backdoors into neural networks during inference, threatening AI integrity.
Generative Design Optimizes Liquid-Cooling Channels for 2.5D and⭐9
Generative design optimizes liquid-cooling channels for 2.5D and 3D packages, enhancing heat management and performance.