I am newbie for Talend and evaluating it, appreciate if you can provide your feedback on the below..
Planning to replace Ab Inito with Talend for Big Data to create Spark jobs on the Hadoop. Thus, at the high level, I would need to find out where Talend can map to existing experience.
- Talend should be able to support complex ETL tasks to curate/model the data to create snowflake and denormalized views. Is there any limitations on typical use-cases to curate big data and then create marts on the Hive Orc/Parquet?
- Talend generates Spark code, if customization is required, how easy to maintain it?
- Is Talent creating optimized code for Spark or optimization is done further?
- Is the Spark code Spark SQL based and is it the latest versions?
- Is the generated Spark code covering checkpoints so that if the job fails, it can continue after the issue is fixed?
- How to implement meta-data driven ETL with Talend for Big Data?
- Anyhing else you would like to mention as limitation, work around, etc..
