SoCC 2024

Meta Info

Homepage: https://acmsocc.org/2024/index.html

Paper list: https://acmsocc.org/2024/schedule.html

Papers

Large Language Models (LLMs)

• LLM inference

  • Queue Management for SLO-Oriented Large Language Model Serving [Paper]

    • UIUC & IBM Research

• LLM training

  • Distributed Training of Large Language Models on AWS Trainium [Paper]

    • AWS

Mixture of Experts (MoEs)

• MoE inference

  • MoEsaic: Shared Mixture of Experts [Paper]

    • IBM Research

GPU Sharing

• KACE: Kernel-Aware Colocation for Efficient GPU Spatial Sharing [Paper]

  • Stony Brook University

Serverless Computing

• On-demand and Parallel Checkpoint/Restore for GPU Applications [Paper]

  • SJTU IPADS & Shanghai Artificial Intelligence Research Institute

  • gCROP: GPU Checkpoint/Restore made On-demand and Parallel

Resource Scheduler

• Scheduler for deep learning training workloads

  • Hops: Fine-grained heterogeneous sensing, efficient and fair Deep Learning cluster scheduling system [Paper]

    • Anhui University & Institute of Artificial Intelligence, Hefei Comprehensive National Science Center

Distributed Training

• Generative Adversarial Networks (GANs)

  • ParaGAN: A Scalable Distributed Training Framework for Generative Adversarial Networks [Paper]

    • NUS