PySparkIntermediateBroken PySpark FixFree

The Backfill Explosion

A pipeline that normally processes one day of data is asked to backfill six months for a regulatory request.

Scenario context

The same code that works daily now runs for many hours, overloads the cluster, and causes repeated failures.

Business requirement

Fix the PySpark code so the pipeline is correct, scalable, and safe to rerun.

Schema

DataFrames depend on the scenario. Assume large production-scale inputs, skewed keys, retries, and partitioned lake storage.

Sample input

Use the code comments and logs to infer the input shape. Focus on the production failure mode, not local toy execution.

Broken logic / code

# Broken: each loop scans the full history and appends output again.
for run_date in backfill_dates:
    df = spark.read.parquet(raw_path)
    daily = df.filter(F.col('event_date') <= run_date)
    daily.write.mode('append').partitionBy('event_date').parquet(output_path)

Logs / error

[Spark] The Backfill Explosion
Stage progress: most tasks finished, one or more tasks are long-running.
Metrics to inspect: shuffle read, spill, skew ratio, executor lost count, file count.
The same code that works daily now runs for many hours, overloads the cluster, and causes
repeated failures.

Expected output / expected logic

Corrected PySpark code or approach should reduce the failure mode, preserve correctness, and include validation/monitoring.

Your attempt

Write the corrected PySpark approach

Think before revealing the answer. A partial but honest attempt is better practice than reading the model solution first.

Saved locally

Interview-style explanation

Now explain your solution as if you are in an interview: symptom, root cause, fix, edge cases, trade-offs, monitoring, and prevention.