AWS re:Invent 2018: [REPEAT 1] A Serverless Journey: AWS Lambda Under the Hood (SRV409-R1)

Channel:Amazon Web Services · Length:59:12 · Published:2018-11-30 · Views:71615

Watch on YouTube

Summary

An AWS re:Invent 2018 session in which Holly Mesrobian and Marc Brooker explain how AWS Lambda is built, including its sandboxing, isolation and scaling architecture. It covers what happens under the hood when serverless code runs.

Why HN recommends it

HN cites it as the authoritative description of Lambda internals and isolation, by the engineers who built it.

You can watch this talk to learn more — HN, 2019

Distributed systems · Cloud, DevOps & reliability

Subtitles (excerpt)

hi my name is Holly miss rubian and I'm the director of engineering for AWS lambda in a little bit I'll be joined my bark Bricker who's a senior principal engineer and serverless mark and I work together at Amazon on lambda today we plan to walk you through some of the key pieces of the lambda architecture and also some of the innovations that we've been working on by the end of this talk you will have a conceptual understanding of the lambda architecture and understand how your code moves through its systems when you call invoke first a little bit about lambda so you'll understand the scale of what we're doing at just three years after general availability AWS lambda already processes trillions of requests across hundreds of thousands of active customers every month lambda is currently available in all 18 AWS regions and as a foundational service we launch in every new region that AWS launches we have a number of customers that are using lambda to build highly available scalable and secure services including thomson reuters whose processing 4,000 requests per second for its product insights analytics platform finra who performs half a trillion validations of stock trades daily for fraud and anomaly detection and zillow who uses lambda in Kinesis to

On Hacker News

Here are a couple videos from reInvent 2018: Jaso talking about DynamoDB internals [LINK] Marc talking about Lambda internals [VIDEO] — appwiz, 2019-06
I believe AWS does share customer information within a device, but with a lot of sandboxing below that. You can watch this talk to learn more: [VIDEO] I think the threat model is basically that you'd need a KVM kernel 0-day + the ability to exploit it, so a point of privilege such as outside of the firecracker sandbox. — staticassertion, 2019-01
AWS SDE here. The short answer is, in order to get your team to adopt something, you need to make the case that it's better for customers (including things like migration costs). If the modern thing is more efficient, is higher availability, increases velocity, and so on then the case can be made. Some specific examples based on things you cite: * For an example of something OSS or "modern" comin… — discodave, 2019-03

Related videos