Hacker News Reader

To be fair, this sounds like an apples to oranges comparison.

Given any "cloud" instace (VM), it's pretty much guaranteed it will be considerably slower than a dedicated server of same size.

Also, it seems you are comparing a local database (single dedicated server) with a clustered database ("few-node cluster").

That said, I totally get the cost-benefit argument and this is also why I started using dedicated servers for my own projects.

View on HN

The point is it's actually impossible to get good performance, even when I scale the nodes up to their highest the raw CPU is a lot lower, and that really matters. Also the cluster has nothing to do with it as this was testing reads and clusters should be basically the same as all they add is a load balance to any one node. Not to mention I tried every combo: just one server vs two or three. Nothing made a difference. The diff was huge at the time, something like 3x.

Reply on news.ycombinator.com