---
title: "Distributed training simulation"
lesson_id: "06"
---

# Distributed training simulation

- **Lesson ID:** 06
- **Goal:** DDP, FSDP, and ZeRO are memory and communication stories. Simulate them before you rent a cluster.
- **Human lesson:** [06-distributed-training-simulation.html](06-distributed-training-simulation.html)
- **Phase:** Phase 3 Training

## Prerequisites

- The previous week's artifact, or a written note if this is week 1.
- Python 3.11+ and a laptop. CUDA is useful from week 5 onward and not required to read.

## Inputs, outputs, and artifacts

- **Inputs:** The replica from the prior week.
- **Outputs:** Core project 4: a DDP/FSDP/ZeRO intuition simulator.
- **Artifacts:** Lesson notes plus the code named in the human page.

## Agent build steps

1. State the week 1 principle: you do not need to pretrain an 800B model to understand the shape of the problem; you do need progressively realistic replicas.
2. Teach the topics on the human page without inventing paper URLs or news claims.
3. Keep starter code in fenced blocks that match the human lesson.
4. End with the assignment and the opinion checkpoint.
5. Use Edge FDE tone: ship, do not sightseeing. Opinion checkpoints matter.

## Constraints

Plain spoken English. No quizzes. No em dashes. No invented metrics, dates, or citations. Brand Edge FDE only. Name well-known papers by title only: Attention Is All You Need, InstructGPT, Direct Preference Optimization, ZeRO, PagedAttention/vLLM, FlashAttention, Scaling Laws for Neural Language Models.

## Key concepts

- DDP replicates the model and all-reduces gradients.
- ZeRO/FSDP shard state so memory per rank falls and communication rises.
- Adam moments are often larger than the fp16 weights.
- Activation memory can dwarf parameter memory.

## Takeaways

- Ship a byte simulator before you book a cluster.
- Name ZeRO as a paper title, not as a mood.
- Write one failure mode you will actually run into.

## Acceptance checks

- Human HTML has Key concepts and Takeaways.
- Opinion checkpoint is present.
- No em dash and no fake URL.
- [ ] Proceed to [lesson 07 brief](07-supervised-fine-tuning.llms.md).