| name | syn-spark-skew |
| description | Operate, configure, troubleshoot, and manage Apache Spark. Use for Apache Spark engineering tasks, day-to-day operations, and production support. |
Apache Spark: data skew and partition tuning
When to use
Use this skill for data skew and partition tuning in Apache Spark. This skill
specifically handles data skew and partition tuning and nothing else in the
Apache Spark family.
Procedure
Diagnose skewed partitions from the UI, repartition/salt hot keys, and tune spark.sql.shuffle.partitions.
Notes
This is the Apache Spark skill dedicated to data skew and partition tuning.
Sibling skills cover other Apache Spark capabilities; this one
is the right choice only when the task is about data skew and partition tuning.