#general

ClickHouse vs Elasticsearch for Small-Scale Applications

TLDR Matt asked if ClickHouse is a better alternative than Elasticsearch for small-scale applications. Ankit shared a link and asked about the scale Matt was working with.

Powered by Struct AI
4
7mo
Solved
Join the chat
Feb 17, 2023 (7 months ago)
Matt
Photo of md5-1c2b014c9d4185f45dd3e08b55130ccb
Matt
03:23 PM
Sorry, I had another question to add to my question above. I understand that SigNoz uses ClickHouse, a columnar database. From what I understand, columnar databases like ClickHouse are better than say Elasticsearch for observability data. But I also have only come across use cases where the performance difference happens at a large enough scale (e.g. Uber and Cloudflare). I'm wondering if ClickHouse is still a better alternative than Elasticsearch for small-scale applications, and if so, why?
Ankit
Photo of md5-dbe7088320fe1d922707613e02f3420d
Ankit
03:39 PM
I do not see 500GB of data as large https://signoz.io/blog/logs-performance-benchmark/
03:40
Ankit
03:40 PM
what is the scale that you work with? Matt
Matt
Photo of md5-1c2b014c9d4185f45dd3e08b55130ccb
Matt
06:31 PM
Thanks for your response Ankit. I'm new to all this and am justing doing research for a school project so I'm not really working with any scale. I just want to know if the performance advantage that ClickHouse offers only happens at the scale of an Uber or CloudFlare. Thanks for the link also, I'll take a look.