Collect the required data from the appropriate data sources and move the data to a single centralized data store in the cloud.

Why?

Because the data might be located in different locations, such as on-premises or in the cloud. The data might also arrive at different speed and intervals.

Steps

  1. Read data from source data store
  2. Perform processing, such as:
    • Serialization/deserialization
    • Compression/decompression
    • Column mapping
  3. Write data to the destination data store (sink)