AI & ML
Storing the Gemma 4 K=V Cache Once: Exact Global-KV Memory Savings in a Paged Serving Engine with Unchanged Attention Kernels
arXiv cs.DC0 views
arXiv:2611.23035v2 Announce Type: replace-cross Abstract: In the global attention layers of Gemma 4 31B, the value is the normalized key projection, and the key is the value scaled by a fixed weight, with rotary position encoding on one quarter of its dimensions. The Gemma 4 report states that this design reduces the global KV cache by up to 37.5%, but the open serving engines that we checked still store a full key and a full value: 1,024 values per KV head and token. We store one 640-value row instead, exact in real arithmetic, and an unchanged paged attention kernel reads it through two overlapping strided views as its key cache and its value cache. In a vLLM prototype with the 16-bit 31B model, the outputs are bit-identical to an emulation of the same arithmetic on the stock cache, 2 of 432 final answers differ from the stock engine (within its run-to-run noise), and the steady decode KV memory per request falls by a factor of 1.34 to 1.40 at 32K to 64K tokens. On a prefill/decode pair of two A100 GPUs with NIXL transfer and CPU offload, a decode engine with a full 15 GiB pool ran 1.38 times the stock decode tokens per step for 16K-token prompts (repeated to 0.06%) and 1.24 times for 8K-token prompts; a model of the vLLM block pool predicted both ratios to within 1% (the engine was 0.7% to 0.9% higher) and matched the change between them. The eager decode throughput rose by 1.30 and 1.11 times, and on this path the compact engine gave the same output tokens as the emulation for 8 prompts at batch 1. The run-time prototype makes an eager step 4% to 9% slower at equal batch, and the method does not yet stack with an 8-bit KV cache at the default scale.
Source:https://arxiv.org/abs/2611.23035PreviousCharacterizing High Bandwidth Flash for LLM ServingNextSafe AI Evaluation: A Systems Perspective
Related posts
Trying to get our Arms around AI efficiency
0Stack Overflow Blog
Why an Embedded Game Fails on Some Phones, and How to Debug It
0dev.to
Microsoft 365 vs Office 2024: The Break-Even Math
0dev.to
Master AI Evals: Build Reliable, Scalable Model Evaluation Pipelines
0dev.to
What to give someone who misses a place
0dev.to
Google is pushing OEMs against GrapheneOS and bootloader unlocking
0Hacker News