After doris upgraded from version 2.1.6 to 4.1.0, the query and insertion efficiency became very low. #66587
Unanswered
yizhimifeng
asked this question in
A - General / Q&A
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
I. Version span
Upgrade from Doris 2.1.6 to 4.1.0
II. Writing efficiency is reduced
The same jdbc [synchronisation] task (using dataworks to write odps data to doris), the original version 2 can basically reach 1w per second, and now the maximum is only 1k per second after version 4? What writing strategy has been updated in the new version?
III. Low query efficiency
A simple query statement will take a long time and even get stuck. This problem is probably due to normal workload. After setting a new workload, the query speed will be faster. Is there a specific explanation for this?
IV. Continuous abnormality of statistical information
FE logs continue to flash a large number of logs. Is the new version more stricter for illegal dates, or is the 2nd version the same? Will this seriously affect the CPU and memory of the be node? Is this related to the reduction in writing and query efficiency?
All reactions