TLDR: Paper Notes - On-demand Container Loading in AWS Lambda
Date: 2024-09-15 Source: https://arpitbhayani.me/blogs/on-demand-container-loading-in-aws-lambda
Overview
Explore AWS Lambda's efficient scaling - block-level loading, deduplication with encryption, & erasure coding for low-latency container startups. These are my notes based on the paper Paper Notes - On-demand Container Loading in AWS Lambda AWS Lambda scales to handle millions of requests per second, provision containers at a rate of 15,000 per second, and achieve cold-start times as low as 50ms, even for large container images (up to 10GiB).
Key Points
- deduplication of blocks with convergent encryption
- use of erasure coding in caching for optimizing tail latencies
- Block-level demand loading: Traditional approaches would require loading entire containers into memory before starting the execution.
- Deduplication with Convergent Encryption: Most containers used across customers have a common base image and other layers.