English(EN) DCC: Data-Centric Compilation of Machine Learning Kernels for Processing-In-Memory Architectures

新编译器DCC优化内存处理架构上的机器学习内核

作者 PulseAugur 编辑部 · [1 个来源] · 2026-05-25 04:00

研究人员开发了DCC，这是一种新颖的以数据为中心的编译器，旨在优化内存处理（PIM）架构上的机器学习内核。该编译器通过联合优化数据重排和计算代码优化这两个相互依赖的过程来应对挑战。DCC通过多层抽象支持多个PIM后端，并已展示出显著的加速效果，与仅使用GPU相比，在HBM-PIM上实现了高达7.68倍的加速，在AttAcc PIM上实现了13.17倍的加速。对于端到端的LLM推理，AttAcc上的DCC将GPT-3和LLaMA-2的平均速度提高了4.52倍。 AI

影响在专用的内存处理硬件上实现LLM推理和其他机器学习工作负载的显著加速。

排序理由该集群包含一篇学术论文，详细介绍了一种用于优化特定硬件架构上机器学习内核的新编译器。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.LG 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 0 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群包含一篇学术论文，详细介绍了一种用于优化特定硬件架构上机器学习内核的新编译器。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, infra; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: 108 days old; Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.LG TIER_1 English(EN) · Peiming Yang, Sankeerth Durvasula, Ivan Fernandez, Mohammad Sadrosadati, Onur Mutlu, Gennady Pekhimenko, Christina Giannoula · 2026-05-25 04:00

DCC：面向内存处理架构的机器学习内核的以数据为中心的编译

arXiv:2511.15503v2 Announce Type: replace-cross Abstract: High-performance Host processors can integrate Processing-In-Memory (PIM) devices, which can accelerate memory-intensive kernels of Machine Learning (ML) models, including Large Language Models (LLMs), by leveraging the la…

本文如何被排名

报道来源 [1]

DCC：面向内存处理架构的机器学习内核的以数据为中心的编译

相关实体

相关话题