[SERVER-70961] extremely slow performance compare to other dbs Created: 28/Oct/22  Updated: 04/Nov/22  Resolved: 04/Nov/22

Status: Closed
Project: Core Server
Component/s: None
Affects Version/s: None
Fix Version/s: None

Type: Bug Priority: Major - P3
Reporter: mosinnik N/A Assignee: Yuan Fang
Resolution: Done Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Issue Links:
Related
Operating System: ALL
Participants:

 Description   

Few week ago I found comparison clickhouse vs other analytics DBs. So there was a mongo with a very bad times. I rewrote queries and add indexes to speed up (from ×3628.01 to ×444.53 current). But in some cases I can't got better results and even got worse (Q30 and Q31) with index usage. Count (Q0) didn't use index or use them in some strange way.
Bench repo: https://github.com/ClickHouse/ClickBench
Comparison site: https://benchmark.clickhouse.com/

So I think something going wrong on core side and could be reworked/optimized if so much time needed to make no so complex queries.
Other queries also need to be checked and extremely slow data loading too (more than 12 hours for 100m docs with 16 indexes)

I also tested timeseries collections but queries was even slower. U can check it anyway by creating TS collections instead of just importing data in benchmark script



 Comments   
Comment by Yuan Fang [ 04/Nov/22 ]

Hi mosinnik@yandex.ru ,

Thank you for your report. I appreciate your efforts in trying to get better results for MongoDB. I believe investigating the result of MongoDB on ClickBench would be valuable for the continuous improvement of MongoDB's performance and we are already evaluating ClickBench internally.

However, this JIRA project (SERVER) is for bug reports, as this ticket looks more like a general improvement request to me, I would encourage you to provide your input to our product team by submitting feedback to our MongoDB Feedback Engine.

Otherwise, for this issue, I'd like to encourage you to start by asking our community for help by posting on the MongoDB Developer Community Forums, if possible, include more details that are helpful for investigation, for instance: queries/indexes used, and the results with/without index usage.

If the discussion there leads you to suspect a bug in the MongoDB server, then we'd want to investigate it as a possible bug here in the SERVER project.

Regards,
Yuan

Generated at Thu Feb 08 06:17:37 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.