# An ingestion pipeline hit a 10x flash-sale spike

Canonical URL: <https://datadriven.io/problems/an-ingestion-pipeline-hit-a-10x-flash-sale-spike-the-produc-25dcbce6>

Domain: Pipeline Design · Difficulty: medium

## Problem

An ingestion pipeline hit a 10x flash-sale spike. The producer writes 50k events/sec; the consumer drains at 5k; there is no durable queue between them. Memory grows on the consumer host until it crashes; the producer keeps writing into a closed socket; events are silently lost. The section names three backpressure strategies: buffer the spike (Kafka, Kinesis, SQS, Pub/Sub), slow the producer (rate-limit at source), or drop selectively (sample or age out). The pick depends on whether data is per-event optional or must not be lost. Place the buffer by adding a durable queue between producer and consumer whose name states the depth and the strategy for events past that depth.

## Related

- [All practice problems](https://datadriven.io/problems)
- [Mock interview mode](https://datadriven.io/interview/an-ingestion-pipeline-hit-a-10x-flash-sale-spike-the-produc-25dcbce6)
- [System Design Interview Questions](https://datadriven.io/data-engineering-system-design)
- [Data Engineering Interview Prep Guide](https://datadriven.io/data-engineer-interview-prep)
- [Daily Challenge](https://datadriven.io/daily)

---

Source: DataDriven (https://datadriven.io). 100% free data engineering interview prep. Live code execution against Postgres 16, Python 3.11, and Spark sandboxes. No paywall, no premium tier, no signup gate.