I'm not sure what the reason is but with 2023.1 there are some errors in %SQL_Diag.Result when loading data from my datasets. The errors are different from 2022.3.
After changing some columns in my data model from NOT NULLABLE to NULLABLE the errors are gone and the performance with IRIS 2023.1 is the same as with 2022.3
I've upgraded my small test project https://github.com/andreas5588/openflights_dataset from 2022.2.0.368.0 to IRIS 2023.1.0.207.0 and I've discovered that the SQL execution time was much slower than expected.
Within this project I load the data using "LOAD DATA FROM FILE" sql statement and it looks like that this is taking up the time.
A "docker-compose build" with IRIS 2022.2.0.368.0 took about about 17 seconds on my machine, the same with IRIS 2023.1.0.207.0 took 300 seconds!
The image with IRIS 2022.2.0.368.0 is 2.87Gb and the image file with IRIS 2023.1.0.207.0 image is 3.21Gb.
This is the only change I've made https://github.com/andreas5588/openflights_dataset.png)
Any idea whats going on?
Andreas
SQL DATA LENS release 3.15 is out! Download here for free
The new AI features available in the world are great for a small development company. It allows me to focus on integrating new features into SQL DATA LENS and let an AI bot generate some promotional videos.
What do you think about this:
or what do you think about that?
.png)