Review cards · 18 cards
Queues
Message queues, pub/sub, delivery guarantees and smoothing load with async work.
Cards
- A job needs a 200 MB video. Why not put the file in the queue message?
- Several workers pulling from one queue, so that each message is handled by only one of them, is the _____ pattern.
- One message makes the consumer fail every time it is processed. What should happen to it?
- Why put a queue between a bursty producer and a slower consumer?
- An "order placed" event must reach the email service, the analytics service and the warehouse service. Which messaging model fits?
- Which metric best tells you a worker fleet should scale up?
- Password-reset emails wait behind a 10-million-message newsletter in the same queue. How do you fix it?
- 3,000 messages arrive per second, and each takes a worker 50 ms to process, one at a time. How many workers keep up?
- Your queue delivers at least once. What must every consumer do?
- After an outage, a queue holds 1.8 million messages. Consumers process 2,000 a second, while producers keep adding 1,500 a second. How long until the backlog is gone?
- What is backpressure, and why is an unbounded queue not a substitute for it?
- Feeds are built by fan-out on write. A celebrity with 50 million followers posts. What do large feed systems do?
- A consumer's call to a provider fails. How do you retry it later without blocking the messages behind it?
- Fan-out on write versus fan-out on read for a feed: what does each cost?
- When does a log like Kafka fit better than a queue like SQS or RabbitMQ?
- Events for each bank account must be processed in order, but the total volume needs hundreds of consumers. What do you do?
- Why must a new post be stored durably before its fan-out is queued?
- In SQS, a message a consumer has received is hidden from other consumers for the _____. If the consumer does not delete it before that ends, the message becomes visible and is _____.
More topics
- Estimation 21 cards
- Networking 16 cards
- API design 17 cards
- Caching 21 cards
- Databases 22 cards
- Replication 15 cards
- Sharding 18 cards
- Consistency 19 cards
- Streaming 18 cards
- Availability 14 cards
- Resilience 16 cards
- Storage 14 cards
- Realtime 15 cards
- Data structures 16 cards
- Security 17 cards
- Observability 18 cards
- Coordination 16 cards