Gradient Dissent: Conversations on AI cover image

Gradient Dissent: Conversations on AI

Advanced AI Accelerators and Processors with Andrew Feldman of Cerebras Systems

Jun 22, 2023

01:00:10

Creator website

Episode guests

AI Summary

AI Chapters

Episode notes

Podcast summary created with Snipd AI

Quick takeaways

Cerebras Systems' large chips enable fast and efficient AI compute, outperforming traditional GPUs and clusters in terms of speed and memory bandwidth utilization.

Data pipeline management and model deployment are crucial aspects that require systematic and methodological approaches to ensure robust and reliable ML model deployment.

Deep dives

Building Big Chips for Fast AI Compute

Andrew Feldman, CEO of SarahBress Systems, discusses the company's approach to building big chips for AI work. They built the largest chip in chip-building history, enabling fast and efficient AI compute. By keeping a significant amount of work on the chip and minimizing data movement, their system achieves high-speed processing with low power consumption. The company's approach allows for the training of large machine learning models, such as GPT, with remarkable performance improvements over GPUs and clusters. Additionally, they address the challenges of managing data pipelines and the debugging process during model deployment.

1.

Introduction

4min

2.

The Downsides of Having One Giant Chip

3min

3.

The History of the GPU

3min

4.

The Importance of Fine-Grained Control in GPUs

2min

5.

How to Make a Chip That Talks to Each Other

2min

6.

The Speed of Light

2min

7.

How to Treat Temperatures as First Class Citizens

3min

8.

The Pain of TPUs in Machine Learning

3min

9.

The Tradeoffs High Torch Made in Compiling Python Code

2min

10.

The Tradeoff Between TensorFlow and PyTorch

3min

11.

How to Add a Little Bit More Data

4min

12.

How to Add a Little Bit More Data

3min

13.

How to Add a Little Bit More Data

3min

14.

How to Add a Little Bit More Data

5min

15.

How to Add a Little Bit More Data

4min

16.

The Nature of Chip Building

47sec

17.

The Future of AI

2min

18.

How to Rent Dedicated Clusters in the Cloud

2min

19.

The Importance of Thoughtful Engineering Leadership

3min

20.

The Hardest Part of the Life Cycle of an ML Model

4min

21.

Apple's Future in ML

3min

On this episode, we’re joined by Andrew Feldman, Founder and CEO of Cerebras Systems. Andrew and the Cerebras team are responsible for building the largest-ever computer chip and the fastest AI-specific processor in the industry.

We discuss:

- The advantages of using large chips for AI work.

- Cerebras Systems’ process for building chips optimized for AI.

- Why traditional GPUs aren’t the optimal machines for AI work.

- Why efficiently distributing computing resources is a significant challenge for AI work.

- How much faster Cerebras Systems’ machines are than other processors on the market.

- Reasons why some ML-specific chip companies fail and what Cerebras does differently.

- Unique challenges for chip makers and hardware companies.

- Cooling and heat-transfer techniques for Cerebras machines.

- How Cerebras approaches building chips that will fit the needs of customers for years to come.

- Why the strategic vision for what data to collect for ML needs more discussion.

Resources:

Andrew Feldman - https://www.linkedin.com/in/andrewdfeldman/

Cerebras Systems - https://www.linkedin.com/company/cerebras-systems/

Cerebras Systems | Website - https://www.cerebras.net/

Thanks for listening to the Gradient Dissent podcast, brought to you by Weights & Biases. If you enjoyed this episode, please leave a review to help get the word out about the show. And be sure to subscribe so you never miss another insightful conversation.

#OCR #DeepLearning #AI #Modeling #ML

Get the Snipd
podcast app

Unlock the knowledge in podcasts with the podcast player of the future.

App store banner

Play store banner

AI-powered
podcast player

Listen to all your favourite podcasts with AI-powered features

Discover
highlights

Listen to the best highlights from the podcasts you love and dive into the full episode

Save any
moment

Hear something you like? Tap your headphones to save it with AI-generated key takeaways

Share
& Export

Send highlights to Twitter, WhatsApp or export them to Notion, Readwise & more

AI-powered
podcast player

Listen to all your favourite podcasts with AI-powered features

Discover
highlights

Listen to the best highlights from the podcasts you love and dive into the full episode