@@ -21,50 +21,163 @@ This directory contains the textbook source and build system for contributors.
2121
2222## What You Will Learn
2323
24- | ML Concepts | Systems Engineering |
25- | -------------| ---------------------|
26- | Neural networks and deep learning | Memory hierarchies and caching |
27- | Model architectures (CNNs, Transformers) | Hardware accelerators (GPUs, TPUs, NPUs) |
28- | Training and optimization | Distributed systems and parallelism |
29- | Inference and deployment | Power and thermal management |
30- | Compression and quantization | Latency, throughput, and efficiency |
24+ <table >
25+ <thead >
26+ <tr>
27+ <th width="50%">ML Concepts</th>
28+ <th width="50%">Systems Engineering</th>
29+ </tr>
30+ </thead >
31+ <tbody >
32+ <tr>
33+ <td>Neural networks and deep learning</td>
34+ <td>Memory hierarchies and caching</td>
35+ </tr>
36+ <tr>
37+ <td>Model architectures (CNNs, Transformers)</td>
38+ <td>Hardware accelerators (GPUs, TPUs, NPUs)</td>
39+ </tr>
40+ <tr>
41+ <td>Training and optimization</td>
42+ <td>Distributed systems and parallelism</td>
43+ </tr>
44+ <tr>
45+ <td>Inference and deployment</td>
46+ <td>Power and thermal management</td>
47+ </tr>
48+ <tr>
49+ <td>Compression and quantization</td>
50+ <td>Latency, throughput, and efficiency</td>
51+ </tr>
52+ </tbody >
53+ </table >
3154
3255### The ML ↔ Systems Bridge
3356
34- | You know... | You will learn... |
35- | -------------| -------------------|
36- | How to train a model | How training scales across GPU clusters |
37- | That quantization shrinks models | How INT8 math maps to silicon |
38- | What a transformer is | Why KV-cache dominates memory |
39- | Models run on GPUs | How schedulers balance latency vs throughput |
40- | Edge devices have limits | How to co-design models and hardware |
57+ <table >
58+ <thead >
59+ <tr>
60+ <th width="50%">You know...</th>
61+ <th width="50%">You will learn...</th>
62+ </tr>
63+ </thead >
64+ <tbody >
65+ <tr>
66+ <td>How to train a model</td>
67+ <td>How training scales across GPU clusters</td>
68+ </tr>
69+ <tr>
70+ <td>That quantization shrinks models</td>
71+ <td>How INT8 math maps to silicon</td>
72+ </tr>
73+ <tr>
74+ <td>What a transformer is</td>
75+ <td>Why KV-cache dominates memory</td>
76+ </tr>
77+ <tr>
78+ <td>Models run on GPUs</td>
79+ <td>How schedulers balance latency vs throughput</td>
80+ </tr>
81+ <tr>
82+ <td>Edge devices have limits</td>
83+ <td>How to co-design models and hardware</td>
84+ </tr>
85+ </tbody >
86+ </table >
4187
4288### Book Structure
4389
4490This textbook is organized into ** two volumes** following the Hennessy & Patterson pedagogical model:
4591
46- | Volume | Theme | Focus |
47- | --------| -------| -------|
48- | ** Volume I** | Build, Optimize, Deploy | Single-machine ML systems, foundational principles |
49- | ** Volume II** | Scale, Distribute, Govern | Distributed systems at production scale |
92+ <table >
93+ <thead >
94+ <tr>
95+ <th width="20%">Volume</th>
96+ <th width="30%">Theme</th>
97+ <th width="50%">Focus</th>
98+ </tr>
99+ </thead >
100+ <tbody >
101+ <tr>
102+ <td><b>Volume I</b></td>
103+ <td>Build, Optimize, Deploy</td>
104+ <td>Single-machine ML systems, foundational principles</td>
105+ </tr>
106+ <tr>
107+ <td><b>Volume II</b></td>
108+ <td>Scale, Distribute, Govern</td>
109+ <td>Distributed systems at production scale</td>
110+ </tr>
111+ </tbody >
112+ </table >
50113
51114#### Volume I: Build, Optimize, Deploy
52115
53- | Part | Focus | Chapters |
54- | ------| -------| ----------|
55- | ** Foundations** | Core concepts | Introduction, ML Systems, DL Primer, Architectures |
56- | ** Development** | Building blocks | Workflow, Data Engineering, Frameworks, Training |
57- | ** Optimization** | Making it fast | Efficient AI, Optimizations, HW Acceleration, Benchmarking |
58- | ** Deployment** | Making it work | Serving, MLOps, Responsible Engineering |
116+ <table >
117+ <thead >
118+ <tr>
119+ <th width="20%">Part</th>
120+ <th width="25%">Focus</th>
121+ <th width="55%">Chapters</th>
122+ </tr>
123+ </thead >
124+ <tbody >
125+ <tr>
126+ <td><b>Foundations</b></td>
127+ <td>Core concepts</td>
128+ <td>Introduction, ML Systems, DL Primer, Architectures</td>
129+ </tr>
130+ <tr>
131+ <td><b>Development</b></td>
132+ <td>Building blocks</td>
133+ <td>Workflow, Data Engineering, Frameworks, Training</td>
134+ </tr>
135+ <tr>
136+ <td><b>Optimization</b></td>
137+ <td>Making it fast</td>
138+ <td>Efficient AI, Optimizations, HW Acceleration, Benchmarking</td>
139+ </tr>
140+ <tr>
141+ <td><b>Deployment</b></td>
142+ <td>Making it work</td>
143+ <td>Serving, MLOps, Responsible Engineering</td>
144+ </tr>
145+ </tbody >
146+ </table >
59147
60148#### Volume II: Scale, Distribute, Govern
61149
62- | Part | Focus | Chapters |
63- | ------| -------| ----------|
64- | ** Foundations of Scale** | Infrastructure | Infrastructure, Storage, Communication |
65- | ** Distributed Systems** | Coordination | Distributed Training, Fault Tolerance, Inference, Edge Intelligence |
66- | ** Production Challenges** | Operations | On-device Learning, Privacy & Security, Robust AI, Ops at Scale |
67- | ** Responsible Deployment** | Trust | Responsible AI, Sustainable AI, AI for Good, Frontiers |
150+ <table >
151+ <thead >
152+ <tr>
153+ <th width="25%">Part</th>
154+ <th width="20%">Focus</th>
155+ <th width="55%">Chapters</th>
156+ </tr>
157+ </thead >
158+ <tbody >
159+ <tr>
160+ <td><b>Foundations of Scale</b></td>
161+ <td>Infrastructure</td>
162+ <td>Infrastructure, Storage, Communication</td>
163+ </tr>
164+ <tr>
165+ <td><b>Distributed Systems</b></td>
166+ <td>Coordination</td>
167+ <td>Distributed Training, Fault Tolerance, Inference, Edge Intelligence</td>
168+ </tr>
169+ <tr>
170+ <td><b>Production Challenges</b></td>
171+ <td>Operations</td>
172+ <td>On-device Learning, Privacy & Security, Robust AI, Ops at Scale</td>
173+ </tr>
174+ <tr>
175+ <td><b>Responsible Deployment</b></td>
176+ <td>Trust</td>
177+ <td>Responsible AI, Sustainable AI, AI for Good, Frontiers</td>
178+ </tr>
179+ </tbody >
180+ </table >
68181
69182---
70183
@@ -143,11 +256,28 @@ book/
143256
144257## Documentation
145258
146- | Audience | Resources |
147- | ----------| -----------|
148- | ** Readers** | [ Online Book] ( https://mlsysbook.ai/book/ ) ・ [ Volume I] ( https://mlsysbook.ai/vol1/ ) ・ [ Volume II] ( https://mlsysbook.ai/vol2/ ) ・ [ PDF] ( https://mlsysbook.ai/book/assets/downloads/Machine-Learning-Systems.pdf ) |
149- | ** Contributors** | [ CONTRIBUTING.md] ( docs/CONTRIBUTING.md ) ・ [ BUILD.md] ( docs/BUILD.md ) |
150- | ** Developers** | [ DEVELOPMENT.md] ( docs/DEVELOPMENT.md ) ・ [ BINDER.md] ( docs/BINDER.md ) |
259+ <table >
260+ <thead >
261+ <tr>
262+ <th width="20%">Audience</th>
263+ <th width="80%">Resources</th>
264+ </tr>
265+ </thead >
266+ <tbody >
267+ <tr>
268+ <td><b>Readers</b></td>
269+ <td><a href="https://mlsysbook.ai/book/">Online Book</a> ・ <a href="https://mlsysbook.ai/vol1/">Volume I</a> ・ <a href="https://mlsysbook.ai/vol2/">Volume II</a> ・ <a href="https://mlsysbook.ai/book/assets/downloads/Machine-Learning-Systems.pdf">PDF</a></td>
270+ </tr>
271+ <tr>
272+ <td><b>Contributors</b></td>
273+ <td><a href="docs/CONTRIBUTING.md">CONTRIBUTING.md</a> ・ <a href="docs/BUILD.md">BUILD.md</a></td>
274+ </tr>
275+ <tr>
276+ <td><b>Developers</b></td>
277+ <td><a href="docs/DEVELOPMENT.md">DEVELOPMENT.md</a> ・ <a href="docs/BINDER.md">BINDER.md</a></td>
278+ </tr>
279+ </tbody >
280+ </table >
151281
152282Binder is the public automation API for book build/validate/maintenance workflows.
153283Use Binder subcommands in editor integrations and CI where possible.
@@ -169,12 +299,32 @@ We welcome contributions! See [docs/CONTRIBUTING.md](docs/CONTRIBUTING.md) for g
169299
170300## Related
171301
172- | Component | Description |
173- | -----------| -------------|
174- | ** [ Main README] ( ../README.md ) ** | Project overview and ecosystem |
175- | ** [ TinyTorch] ( ../tinytorch/ ) ** | Build ML frameworks from scratch |
176- | ** [ Hardware Kits] ( ../kits/ ) ** | Deploy to Arduino, Raspberry Pi, edge devices |
177- | ** [ Website] ( https://mlsysbook.ai/book/ ) ** | Read the book online |
302+ <table >
303+ <thead >
304+ <tr>
305+ <th width="30%">Component</th>
306+ <th width="70%">Description</th>
307+ </tr>
308+ </thead >
309+ <tbody >
310+ <tr>
311+ <td><b><a href="../README.md">Main README</a></b></td>
312+ <td>Project overview and ecosystem</td>
313+ </tr>
314+ <tr>
315+ <td><b><a href="../tinytorch/">TinyTorch</a></b></td>
316+ <td>Build ML frameworks from scratch</td>
317+ </tr>
318+ <tr>
319+ <td><b><a href="../kits/">Hardware Kits</a></b></td>
320+ <td>Deploy to Arduino, Raspberry Pi, edge devices</td>
321+ </tr>
322+ <tr>
323+ <td><b><a href="https://mlsysbook.ai/book/">Website</a></b></td>
324+ <td>Read the book online</td>
325+ </tr>
326+ </tbody >
327+ </table >
178328
179329---
180330
0 commit comments