Hacker News new | ask | show | jobs
by pier25 1897 days ago
I've been using Fauna for a year or so.

What's slowing Fauna here are the global writes and the non-idiomatic FQL queries.

Right now the code is doing a bunch of separate queries, but in idiomatic FQL this would be done in a single transaction.

Edit:

I'm going to do a PR to update the FQL code if the author accepts it.
1 comments
noahfschr 1897 days ago
Sure, I will.
link
pier25 1897 days ago
Thanks for the merge. As you pointed in the PR you're only measuring the latency of the first query so it will have no effect in the benchmark. I'm guessing the "current request" latency will improve though, no?

I have to say the latencies you're getting are much higher that I've experienced with Fauna. Obviously it's expected for a KV database to be faster, but I'd be surprised to get more than 100ms of 50th percentile.

link
noahfschr 1897 days ago
Current request latency is also the read latency (from lambda function -> db). I should have been more clear on that.
link
pier25 1897 days ago
Still, almost a second of total latency (50th percentile) is super high.

I'd love to do the same test from Cloudflare Workers instead of AWS Lambda. Could you please DM on Twitter to discuss this?

https://twitter.com/PierB

Edit:

I just saw this comment by Evan from Fauna which explains why the latency could be so high:

https://news.ycombinator.com/item?id=26806541

link