Serverless Clickhouse with Tyler Wells
Software Huddle
00:00
Evolution of Data Ingestion and Processing Methods
The chapter explores the evolution of data ingestion methods in Propel, discussing the initial emphasis on Snowflake integration, the development of additional methods like webhooks and S3 in Parquet format, and plans to support platforms like Kafka. It delves into Clickhouse's efficiency in handling data ingestion, discussing its capabilities with S3 and Parquet files, web hook strategies for handling JSON data, and the challenges faced in processing modern data infrastructure. The conversation also covers the importance of materialized views in Clickhouse, access policies for multi-tenant scenarios, and considerations around data pooling based on different use cases.
Transcript
Play full episode
