Model Parallelism: Building and Deploying Large Neural Networks

Instructor-Led Workshop

Very large deep neural networks (DNNs), whether applied to natural language processing (e.g., GPT-3), computer vision (e.g., huge Vision Transformers), or speech AI (e.g., Wave2Vec 2) have certain properties that set them apart from their smaller counterparts. As DNNs become larger and are trained on progressively larger datasets, they can adapt to new tasks with just a handful of training examples, accelerating the route toward general artificial intelligence. Training models that contain tens to hundreds of billions of parameters on vast datasets isn’t trivial and requires a unique combination of AI, high-performance computing (HPC), and systems knowledge. The goal of this course is to demonstrate how to train the largest of neural networks and deploy them to production.

Learning Objectives

By participating in this workshop, you’ll learn how to::

  • Train neural networks across multiple servers.
  • Use techniques such as activation checkpointing, gradient accumulation, and various forms of model parallelism to overcome the challenges associated with large-model memory footprint.
  • Capture and understand training performance characteristics to optimize model architecture.
  • Deploy very large multi-GPU models to production using NVIDIA Triton™ Inference Server.

Workshop Outline

(15 mins)
Introduction to Training of Large Models
(120 mins)
  • Learn about the motivation behind and key challenges of training large models.
  • Get an overview of the basic techniques and tools needed for large-scale training.
  • Get an introduction to distributed training and the Slurm job scheduler.
  • Train a Megatron-LM-based GPT model using data parallelism.
  • Profile the training process and understand execution performance.
Model Parallelism: Advanced Topics
(120 mins)
  • Increase the model size using a range of memory-saving techniques.
  • Get an introduction to tensor and pipeline parallelism.
  • Go beyond natural language processing and get an introduction to DeepSpeed.
  • Auto-tune model performance.
  • Learn about mixture-of-experts models.
Inference of Large Models
(120 mins)
  • Understand the challenges of deployment associated with large models.
  • Explore techniques for model reduction.
  • Learn how to use NVIDIA® TensorRT™ and Faster Transformer libraries.
  • Learn how to use Triton Inference Server.
  • Understand the process of deploying GPT checkpoint to production.
  • See an example of prompt engineering.
Final Review
(15 mins)
  • Review key learnings and answer questions.
  • Complete the assessment and earn a certificate.
  • Complete the workshop survey.
  • Learn how to set up your own AI application development environment.

Start of workshop: 9:00 am. Breaks will be allocated by the instructor, including one hour for lunch. Workshop finishes by 5:30 pm.

Workshop Details

Duration: 8 hours


Technologies: PyTorch, Megatron-LM, DeepSpeed, Slurm, Triton Inference Server, NVIDIA Nsight™

Certificate: Upon successful completion of the assessment, participants will receive an NVIDIA DLI certificate to recognize their subject matter competency and support professional career growth.

Hardware Requirements: Laptop computer capable of running the latest version of Chrome or Firefox. Each participant will be provided with dedicated access to a fully configured, GPU-accelerated workstation in the cloud.

Language: English