Thanks to the folks who joined us live today!
Here's a quick recap of things that were discussed:
• File Stream Interface Development: Discussion on creating a file stream interface for various formats like CSV, XML, and Excel.
• Data Lake Architecture Insights: Hal shared his experiences with target JSON S3, issues with existing target-athena, and potential integration of target iceberg.
• Snowflake Loading Challenges: We gave Hoang a few recommendations for with loading large fields into Snowflake and proposed workarounds using stream maps.
• Log-based MySQL replication: Workarounds were discussed to avoid losing data when binlog files are dropped.