I think the DB layer is probably the least optimized area of dev.to right now. We’re smart enough to keep things simple, but we need to improve so we can hang our hat on this part of the work.
We overcome some naive queries with caching. But it’s not ideal. We may have some budget to hire more in this area, but I think it’s important to personally improve here as well.
I think the DB layer is probably the least optimized area of dev.to right now. We’re smart enough to keep things simple, but we need to improve so we can hang our hat on this part of the work.
We overcome some naive queries with caching. But it’s not ideal. We may have some budget to hire more in this area, but I think it’s important to personally improve here as well.
My education is generally motivated by need. 😄
Same here :D
There are a few quick wins you can try on dev.to's DB, like checking the slow query log, usually you find underused or missing indexes.
A very helpful tool is heroku-pg-extras, it can tell you about locks, outliers, unused indexes and "bloat"
I think learning by need is the best strategy. If no need is there, the brain runs 'D' of its own CRUD operations.