Architecture Thinking Patterns

"It Takes Too Long for Serverless" — Usually Doesn't

The 15-min limit is a signal the problem needs rethinking — not the infrastructure scaling up.

❌ Container Thinking

"This job takes X hours → we need a long-running container / VM / ACI to hold it."

✓ Serverless Thinking

"Why does it take X hours? Move to write-time, fan-out, or checkpoint through a queue."

Click a scenario →
The 3 patterns that eliminate long-running jobs
01
Shift to Write-Time
Don't aggregate at read-time. Update running totals on every write. Reports become instant reads.
Tax YTD, inventory, analytics dashboards
02
Fan-Out Parallelism
If a job has N independent units, run N functions in parallel. Wall time = time for one unit.
Report metrics, bulk emails, doc processing
03
Queue Checkpointing
Each function processes one chunk and enqueues the next. Failures retry only their chunk.
Data migrations, ETL, sequential workflows