Get the latest tech news
Erasure Coding for Distributed Systems
An overview of erasure coding, its trade-offs, and applications in distributed storage systems.
Infrequent access means issuing more IO operations per second won’t be a problematic tax, and the storage savings are significant when compared to storing multiple full copies of every file. I’ll suggest "Practical Performance Evaluation of Space Optimal Erasure Codes for High Speed Data Storage Systems"[7] as the one I liked the most, from which the following figure was taken: [7]: Rui Chen and Lihao Xu. (And if you continue, take what follows with a large grain of salt, as efficient erasure coding is a subject folk have spent years on, and the below is what I’ve collected from a couple of days of reading through papers I only half understand.)
Or read this on Hacker News
