Marc's Blog

About Me

My name is Marc Brooker. I've been writing code, reading code, and living vicariously through computers for as long as I can remember. I like to build things that work. I also dabble in machining, welding, cooking and skiing.

I'm currently an engineer at Amazon Web Services (AWS) in Seattle, where I work on databases, serverless, and serverless databases. Before that, I worked on EC2 and EBS.
All opinions are my own.


My Publications and Videos
@marcbrooker on Mastodon @MarcJBrooker on Twitter

Serial, Parallel, and Quorum Latencies

Why are they letting me write Javascript?

I’ve written before about the latency effects of series (do X, then Y), parallel (do X and Y, wait for them both), and quorum (do X, Y and Z, return when two of them are done) systems. The effects of these different approaches to doing multiple things are quite intuitive. What may not be intuitive, though, is the impact of quorums, and how much quorums can reduce tail latency.

So I put together this little toy simulator.

The knobs are:

So, for example, a traditional 3-of-5 Paxos system would have serial=1, parallel=5, and quorum=3. A length-3 chain replication system would have serial=3, parallel=1, quorum=1. The per-node service time distribution is (for now) assumed to be exponentially distributed with mean 1.

Examples to Try

Have fun!