For SQL engines, simply create tables and views in Okera. The result is a seamless and completely transparent integration to end-user tools and provides all the analytic functionality of the underlying data store. Files in various formats, such as JSON or Parquet, are made accessible in storage backends such as S3 without the need to expose the details to data consumers.
For engines and processing systems, such as Apache Spark and Python, you have the option of creating schemas dynamically using the provided API endpoints of the Okera services.
After creating the necessary schemas, you then proceed to access data through various common interfaces (JDBC/ODBC, REST API).