Introduction
EVL originally stood for Extract–Validate–Load, but it’s now a fully featured ETL (Extract–Transform–Load) tool.
EVL is designed with the Unix philosophies of interoperability and “do one thing, and do it well” in mind.
Templates, a high level of abstraction, and the ability to dynamically create jobs, make for a powerful ETL tool.
Characteristics
- Versatile, i.e. cooperate with other components of the customer’s solution, solving only particular problem.
- High performance, written in C++.
- Lightweight, just install a rpm/deb package or unzip tgz.
- Highly efficient development due to strict command-line approach.
- Managed access to the source code (e.g.
git
). - Linux only, using the best of the system.
- Graphical User Interface – EVL Manager.
Features
- Natively read/write1:
- File formats: CSV, JSON, XML, XLS, XLSX, Parquet, Avro, QVD/QVX, ASN.1
- DBMS: MariaDB/MySQL, Oracle, PostgreSQL, SQLite, ODBC, (near future: Snowflake, Redshift)
- Cloud storages: Amazon S3 and Google Storage
- Hadoop: read/write HDFS, resolve, build and run Spark jobs, Impala/Hive queries
- Partitioning, to partition data and/or parallelize processing
- Productivity boosters, to generate jobs/workflows from metadata
For the most recent information about EVL and supported formats and DBs please check https://www.evltool.com.