Debugging Production-Scale Spark Pipelines

Be able to design and run parallel, branch-isolated Spark experiments to catch and fix bugs that only appear at full production scale, including data skew, collect_list memory bloat, and AQE broadcast join pitfalls.

4 sections ยท 9 lessons

Course outline

Why Scale-Only Bugs Slip Past Small-Sample Testing

  1. Data skew and why 1% samples miss it
  2. When you must test against full, unsampled production data

Building a Parallel Experimentation Loop

  1. Branch-driven isolation for concurrent experiments
  2. Using tmux for session persistence on remote dev boxes
  3. Delegating experiment ops to an AI agent

Spark Memory Internals That Bite at Scale

  1. collect_list and per-row memory bloat
  2. How AQE broadcast joins can silently pin executor memory

Turning Tuning Into Lasting Fixes

  1. Keeping a single experiment log to isolate variables
  2. Why tuning buys headroom, not a new ceiling

Start learning with Wondering