Hacker News Reader

Comment by drchaim | Hacker News Reader

I discovered ClickHouse around 2017-18 and built a PoC to replace Elasticsearch: 5x better storage and qps, in a couple of weeks.

Managers rejected it because it wasn't well known and was seen as "some database made by Russians."

On a personal level, it's quite sad to have seen that train coming so early and not been able to get on board.

View on HN

I had the same experience recently. Turns out ClickHouse would reduce our DB operations by 60%, remove the need for a TSDB, and reduce query times from ~300-500ms (and sometimes ~3s) to roughly ~75ms. Lastly, and most impressively we were already seeing a ridiculous level of compression and our storage cost benchmarks were reduced to the cost of S3. This took a $2-3M storage layer down to one measured in the single thousands per month.

ClickHouse is no panacea but if you understand how your data is accessed and thus how to arrange it you will get so many miles out of it.

Same we are also stuck with ES wish could migrate to clickhouse but not able to do so because of the legacy load.

What do you not like about ES?

Were you using it for simple grep search or actually required advanced searching for eg: BM25. Clickhouse will only help you with grep like search from what I understand.

Actually, there was no search, only on-the-fly aggregations/filtering over "big data". ES was kind of famous at the time, although not the best tool for that job.

afaik CH introduced FTS rececently.

Can clickhouse to search? If not why did you seek to replace elastic with it

Yes https://clickhouse.com/blog/clickhouse-full-text-search-obje...

Thanks

Supply chain risk.

Reply on news.ycombinator.com