paper-with-me

Papers

MAS-Attention: Memory-Aware Stream Processing for Attention Acceleration on Resource-Constrained Edge Devices

2024-11-20 · Mohammadali Shakerdargah, Shan Lu, Chao GAO, Di Niu

The advent of foundation models have revolutionized various fields, enabling unprecedented task accuracy and flexibility in computational linguistics, computer vision and other domains. Attention mechanism has become an essential component of foundation models, due to their superb capability of capturing correlations in a sequence. However, attention results in quadratic complexity in memory and compute as the context length grows. Although many fusion-based exact attention acceleration algorithms have been developed for datacenter-grade GPUs and accelerators leveraging multi-core parallelism and data locality, yet it remains a significant challenge to accelerate attention on resource-constrained edge neural accelerators with limited compute units and stringent on-chip caches. In this paper, we propose a scheme for exact attention inference acceleration on memory-constrained edge accelerators, by parallelizing the utilization of heterogeneous compute units, i.e., vector processing units and matrix processing units. Our method involves scheduling workloads onto these different compute units in a multi-tiered tiling scheme to process tiled vector workloads and matrix workloads in attention as two streams, respecting the workload dependencies. We search for tiling factors to maximize the parallelization of both compute units while considering I/O overhead, and propose a proactive cache overwrite strategy to avoid undesirable cache spills in reality. Extensive results based on open-sourced simulation frameworks show up to 2.75x speedup and 54% reduction in energy consumption as compared to the state-of-the-art attention fusion method (FLAT) in the edge computing scenario. Further experiments on a real-world edge neural processing unit demonstrate speedup of up to 1.76x for attention as compared to FLAT, without affecting model output accuracy.

📄 PDF Abstract BibTeX arXiv:2411.17720

Code (0)

등록된 구현이 없습니다.

Tasks

Edge-computingScheduling

Methods 이 논문이 사용한 방법론

Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Attention 설명 없음

Similar Papers 제목 키워드 기반

Exploring RWKV for Memory Efficient and Low Latency Streaming ASR

2023-09-26 · Keyu An, Shiliang Zhang

Recently, self-attention-based transformers and conformers have been introduced as alternatives to RNNs for ASR acoustic modeling. Nevertheless, the full-sequence attention mechanism is non-streamable and computationally…

Chunking

Attention Itself Could Retrieve.RetrieveVGGT: Training-Free Long Context Streaming 3D Reconstruction via Query-Key Similarity Retrieval

2026-05-10 · Zichen Zou, Xiaosong Jia, Zuxuan Wu, Yu-Gang Jiang arxiv

Visual Geometry Grounded Transformer (VGGT) advances 3D reconstruction via scalable Transformer architecture, but the quadratic complexity of global attention prevents long context application. StreamVGGT enables streami…

3D Reconstruction

DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Heads

2024-10-14 · Guangxuan Xiao, Jiaming Tang, Jingwei Zuo, Junxian Guo 외

Deploying long-context large language models (LLMs) is essential but poses significant computational and memory challenges. Caching all Key and Value (KV) states across all attention heads consumes substantial memory. Ex…

GPUQuantizationRetrieval

Lean Attention: Hardware-Aware Scalable Attention Mechanism for the Decode-Phase of Transformers

2024-05-17 · Rya Sanovar, Srikant Bharadwaj, Renee St. Amant, Victor Rühle 외

Transformer-based models have emerged as one of the most widely used architectures for natural language processing, natural language generation, and image generation. The size of the state-of-the-art models has increased…

Image GenerationText Generation

PAT: Accelerating LLM Decoding via Prefix-Aware Attention with Resource Efficient Multi-Tile Kernel

2025-11-27 · Jinjun Yi, Zhixin Zhao, Yitao Hu, Ke Yan 외 arxiv

LLM serving is increasingly dominated by decode attention, which is a memory-bound operation due to massive KV cache loading from global memory. Meanwhile, real-world workloads exhibit substantial, hierarchical shared pr…