Semiconductor Engineering

Dynamic KV Cache Scheduling in Heterogeneous Memory Systems for LLM Inference (Rensselaer Polytechnic Institute, IBM)

A new technical paper titled “Accelerating LLM Inference via Dynamic KV Cache Placement in Heterogeneous Memory System” was published by researchers at Rensselaer Polytechnic Institute and IBM. “Large ...
Semiconductor Engineering

Arithmetic Intensity In Decoding: A Hardware-Efficient Perspective (Princeton University)

“LLM decoding is bottlenecked for large batches and long contexts by loading the key-value (KV) cache from high-bandwidth memory, which inflates per-token latency, while the sequential nature of ...
InfoWorld

Snowflake open sources SwiftKV to reduce inference workload costs

SwiftKV optimizations developed and integrated into vLLM can improve LLM inference throughput by up to 50%, the company said. Cloud-based data warehouse company Snowflake has open-sourced a new ...
insideHPC

DDN Takes on GPU Waste with KV Cache Performance for AI Reasoning

CHATSWORTH, Calif. — July 18, 2025 DDN today unveiled performance benchmarks that the company said demonstrates how its AI-optimized DDN Infinia platform eliminates GPU waste and delivers the fastest ...
VentureBeat

How attention offloading reduces the costs of LLM inference at scale

Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn More Rearranging the computations and hardware used to serve large language ...