New KV cache compaction technique cuts LLM memory 50x without accuracy loss

MIT researchers developed Attention Matching, a KV cache compaction technique that compresses LLM memory by 50x in seconds — ...

Nvidia shrinks LLM memory 20x without changing model weights

Nvidia's KV Cache Transform Coding (KVTC) compresses LLM key-value cache by 20x without model changes, cutting GPU memory costs and time-to-first-token by up to 8x for multi-turn AI applications.
TMCnet

Penguin Solutions Introduces Industry's First Production-Ready CXL-Based KV Cache Server

Accelerating memory-dependent AI processes, Penguin's MemoryAI KV cache server increases memory capacity by integrating 3 TB of DDR5 main memory and up to eight 1 TB CXL Add-in Cards (AICs). Penguin ...
EDN

Last-level cache has become a critical SoC design element

LLC, positioned between external memory and internal subsystems, stores frequently accessed data close to compute resources.
EDN

CXL initiative tackles memory challenges in heterogeneous computing

In the world of computing, one of the unexpected things to marvel at is the rapid adoption of artificial intelligence (AI) and cloud computing in data centers. These and other forces are driving ...
PC World

How does CPU memory cache work?

In the eighties, computer processors became faster and faster, while memory access times stagnated and hindered additional performance increases. Something had to be done to speed up memory access and ...
CNET

The end of battery-backed cache?

Adaptec has announced a RAID controller series that uses NAND and Supercapacitors to protect data in cache in case of failure. Will Adaptec stand alone? John, a senior partner at Evaluator Group, has ...