paper-with-me

Papers

MARS: Exploiting Multi-Level Parallelism for DNN Workloads on Adaptive Multi-Accelerator Systems

2023-07-23 · Guan Shen, Jieru Zhao, Zeke Wang, Zhe Lin, Wenchao Ding, Chentao Wu, Quan Chen, Minyi Guo

Along with the fast evolution of deep neural networks, the hardware system is also developing rapidly. As a promising solution achieving high scalability and low manufacturing cost, multi-accelerator systems widely exist in data centers, cloud platforms, and SoCs. Thus, a challenging problem arises in multi-accelerator systems: selecting a proper combination of accelerators from available designs and searching for efficient DNN mapping strategies. To this end, we propose MARS, a novel mapping framework that can perform computation-aware accelerator selection, and apply communication-aware sharding strategies to maximize parallelism. Experimental results show that MARS can achieve 32.2% latency reduction on average for typical DNN workloads compared to the baseline, and 59.4% latency reduction on heterogeneous models compared to the corresponding state-of-the-art method.

📄 PDF Abstract BibTeX arXiv:2307.12234

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

TetriServe: Efficiently Serving Mixed DiT Workloads

2025-10-02 · Runyu Lu, Shiqi He, Wenxuan Tan, Shenggui Li 외 arxiv

Diffusion Transformer (DiT) models excel at generating high-quality images through iterative denoising steps, but serving them under strict Service Level Objectives (SLOs) is challenging due to their high computational c…

Image Generation

Exploiting Student Parallelism for Low-latency GPU Inference of BERT-like Models in Online Services

2024-08-22 · Weiyan Wang, Yilun Jin, Yiming Zhang, Victor Junqiu Wei 외

Due to high accuracy, BERT-like models have been widely adopted by discriminative text mining and web searching. However, large BERT-like models suffer from inefficient online inference, as they face the following two pr…

GPU

MarsRL: Advancing Multi-Agent Reasoning System via Reinforcement Learning with Agentic Pipeline Parallelism

2025-11-14 · Shulin Liu, Dong Du, Tao Yang, Yang Li 외 arxiv

Recent progress in large language models (LLMs) has been propelled by reinforcement learning with verifiable rewards (RLVR) and test-time scaling. However, the limited output length of LLMs constrains the depth of reason…

Reinforcement Learning

AlpaServe: Statistical Multiplexing with Model Parallelism for Deep Learning Serving

2023-02-22 · Zhuohan Li, Lianmin Zheng, Yinmin Zhong, Vincent Liu 외

Model parallelism is conventionally viewed as a method to scale a single large deep learning model beyond the memory limits of a single device. In this paper, we demonstrate that model parallelism can be additionally use…

Deep Learning

Polystore++: Accelerated Polystore System for Heterogeneous Workloads

2019-05-24 · Rekha Singhal, Nathan Zhang, Luigi Nardi, Muhammad Shahbaz 외

Modern real-time business analytic consist of heterogeneous workloads (e.g, database queries, graph processing, and machine learning). These analytic applications need programming environments that can capture all aspect…