Performance Tuning

Performance is the cornerstone of any Big Data Processing. Fire is extremely optimized for best performance. Each of the Processors are written for extreme performance, the engine is optimized for the best performance.

There are certain things which need to be taken into account for any Spark job. Fire makes it extremely easy to apply them to a Workflow.

• Caching Level
  • When to use Caching
• Executor Memory, vcores
• Repartioning
  • When saving to JDBC/File etc.
• Debug Mode