Q: What did your data warehousing and BI architecture look like before Snowflake? A: All of those 1.2 billion web events sat in our cloud environment, Azure Blob, as JSON structures. We needed to push those JSON structures to our on-premises SQL server and use its auditing technology to pull those events into a staging area. We set up intraday stored procedures to parse those data point structures into staging tables and integrate them in the holding structure, where facts and dimensions were combined. Then we would take all 1.2 billion events and aggregate them into our SQL Server Analysis Services (SSAS) Tabular structure for analysis in Microsoft Excel. Q: What were your pain points at the time? A: SQL was very slow ingesting JSON. Pulling in all those events and parsing th

Join for free to read