ReadMeFeed / Community knowledge
Real questions. Useful conversations. Find the people who know your stack.
Pulsar's Key_Shared subscription preserves per-key ordering while scaling consumers horizontally. Here's how it routes messages, a working example, and the hot-key and rebalancing trade-offs to design around.
Pulsar's Key_Shared subscription keeps every message key in order while the consumer group scales. How the hash routing works, what hot keys and rebalances cost, and how to verify it on your cluster.
Redis Streams provide durable, ordered messaging with consumer‑group semantics. Learn the minimal design, trust boundaries, ops checks, and when to reconsider.
Azure Cosmos DB Change Feed provides a persistent, ordered log of inserts and updates that microservices can consume directly. This post shows how to set it up, run a sample processor, and consider its limits.
When you need to replay only the most recent value for each key in a Kafka topic, log compaction is the right tool. This guide explains how compaction works, shows a practical example, and discusses trade‑offs and limits so you can decide if it fits your use case.
LinkedIn moved from batch ETL to a Kafka event backbone to keep feeds fresh. Partitioned topics enable independent consumers and incremental updates, with trade-offs in key design and consumer lag.
pulsar
redis
apachekafka
linkedin