iCreatAI News
Skip to content
Trending storyUpdating

Splice-Core hybrid GEMM accelerator paper published

1 reports1 sources18 hr agoUpdated

Understand this story

AI overview

Prabhu Vellaisamy and co-authors have published a paper proposing a temporal unary/binary hybrid GEMM accelerator unit called Splice-Core, which reduces energy consumption by minimizing low-order accumulation latency and retaining value-dependent execution only for high-order activation bits. The paper reports that its two PE variants, Splice-BP and Splice-LUT, achieve geometric-mean tile energy reductions of 77.4% and 76.4% respectively on 26 INT4 CNN and LLM workloads in a 45nm process. These reduction figures are the results reported by the authors.

AI overview based on source reports · Updated 18 hr ago

Coverage timeline

Follow the coverage to explore different perspectives.

Nov 21
  1. Hacker News AI
    Splice-Core: Low-Order-Bit-Splicing GEMM for Energy-Efficient AI Accelerators

    Splice-Core is a temporal-unary/binary (tub) hybrid GEMM unit that minimizes low-order accumulation latency while retaining value-dependent execution for high-order activation bits, with two PE variants: Splice-BP using bit planes and Splice-LUT selecting a shared weight multiple in a single cycle. Compared to a binary systolic array in 45nm across 26 INT4 CNN and LLM workloads, Splice-BP/Splice-LUT achieve 77.4%/76.4% geometric-mean tile energy reductions, and Splice-LUT further raises geometric-mean tile throughput by 44.4% and reduces geometric-mean EDP by 6.13x. Across evaluated W8A8 convolution and expert-projection workloads, Splice-BP and Splice-LUT reduce tile energy by 77.3-88.1%, and post-place-and-route INT4 yields 14%/9% smaller area for Splice-BP/Splice-LUT versus the systolic array.

Activity for this story

Current comparable activity
6
Peak comparable activity
10Nov 21 14:00
Change in the past 24 hours
–

Trends compare the same fully observed participants, which may cover fewer participants than current activity. Move the pointer or tap the chart to inspect hourly activity; use the left and right arrow keys with a keyboard.