I am wary of user reports that claim product X was lousy for them, then they moved to product Y and everything was awesome. Sometimes this means that product X was lousy -- in general or for their use case. Other times it means the team using product X did a lousy job deploying it. It is hard for the reader to figure this out. It can also be hard for some authors to figure this out thanks to the Dunning-Kruger effect so lousy reports will continue to be published. These reports are not my favorite form of marketing and some of the bad ones linger for years. We deserve better especially in the open-source database market where remarkable progress is being made.
I have written before on benchmarketing. Other posts that mention it are here.
Subscribe to:
Post Comments (Atom)
Postgres 18 beta1: small server, IO-bound Insert Benchmark (v2)
This is my second attempt at an IO-bound Insert Benchmark results with a small server. The first attempt is here and has been deprecated b...
-
If you use RocksDB and want to avoid OOM then use jemalloc or tcmalloc and avoid glibc malloc. That was true in 2015 and remains true in 202...
-
I need stable performance from the servers I use for benchmarks. I also need servers that don't run too hot because too-hot servers caus...
-
This provides additional results for Postgres versions 11 through 16 vs Sysbench on a medium server. My previous post is here . The goal is ...
No comments:
Post a Comment