YoVDO

ML Training Acceleration with Heterogeneous Resources in ByteDance

Offered By: CNCF [Cloud Native Computing Foundation] via YouTube

Tags

Machine Learning Courses Kubernetes Courses Distributed Systems Courses GPU Acceleration Courses Heterogeneous Computing Courses RDMA Courses

Course Description

Overview

Save Big on Coursera Plus. 7,000+ courses at $160 off. Limited Time Only!
Explore machine learning training acceleration techniques using heterogeneous resources at ByteDance in this 19-minute conference talk from KubeCon + CloudNativeCon Europe 2022. Delve into strategies for maximizing GPU utilization through sharing mechanisms, optimizing resource allocation with NUMA affinity, and implementing high-throughput network communication using RDMA CNI and Intel SRIOV technology. Gain insights into empowering model training, enhancing performance for large-scale distributed models, and effectively managing diverse CPU/GPU resources. Cover topics including GPU offline training for network and scheduling, GPU online serving, unified GPU scheduling, and future developments in the field.

Syllabus

Intro
GPU Offline Training (Network)
GPU Offline Training (Scheduling).
GPU Online Serving
GPU Unified Scheduling
Future Work


Taught by

CNCF [Cloud Native Computing Foundation]

Related Courses

Future of Computing - IBM Power 9 and beyond
openHPI
SIGCOMM 2020 - Reducto - On-Camera Filtering for Resource-Efficient Real-Time Video Analytics
Association for Computing Machinery (ACM) via YouTube
Offload Annotations - Bringing Heterogeneous Computing to Existing Libraries and Workloads
USENIX via YouTube
Supercomputing Spotlights - Supercomputing Software for Moore and Beyond
Society for Industrial and Applied Mathematics via YouTube
Liquid Metal - Taming Heterogeneity
GOTO Conferences via YouTube