ETL vs ELT Architectures for Continuous Streams

For decades, ETL with products such as Informatica Power Center was the primary method for preparing data for analytics before loading it into a data warehouse. But Big Data and Fast Data changed everything. Traditional ETL platforms were too slow to provide needed data freshness and scale. 

ELT in a cloud data warehouse like Snowflake or Redshift emerged as an attractive alternative. We could simply push the data into the warehouse and then leverage massive compute resources to prepare it for use. But continuously-running transformations in a data warehouse is resource-intensive and expensive, since they weren’t architected or priced for this use case. 

In In this webinar, we will:

• Unpack the pros and cons of these two design patterns and the compromises each requires in terms of performance, cost, ease, flexibility and other factors. 
• Discuss how these factors can be weighed to determine where each pattern makes the most sense for a given use case.
• Show how using Upsolver SQLake eliminates many of these compromises by combining simplicity with scalable performance.

We will demo how you can build ETL pipelines to ingest prepared data into your cloud as well as ELT pipelines where data is transformed and queried from your cloud data lake. Save your seat now!