More accurate calculation of issues in data products

Issue-centric calculation for data products

We evolved the display of issues to be failure-centric rather than asset-centric.

This improves our accuracy when tracking issues that impact data products.

For example, if 2 assets in your data product failed and each had 2 failures, we showed 2 issues.

Now, we add all the failures and give you the total number, in our example, 4.

Redshift query logs

We can now fetch query logs from Redshift. What that means is we can identify cost optimization opportunities for our customers using Redshift.

We can for example:

• See which models run more often that they should
• Which models are not used downstream
• Which models are the most expensive