Load Shedding: Dropping Traffic to Stay Alive
Load shedding explained: overload survival by dropping work deliberately: graceful degradation ladders chosen in advance, server overload protection…
Category
Reliability
Reliability engineering is the practice of keeping systems available and correct, and recovering quickly when they fail. It covers redundancy and failover for high availability, disaster recovery planned around RTO and RPO targets, safe deployment strategies such as blue-green and canary releases, and incident response. Site reliability engineering, as Google defined it, adds error budgets that balance shipping speed against stability.
16 articles
-
-
Idempotency: Designing Operations That Survive Retries
Idempotency explained: operations that survive retries and duplicates, the idempotency key from client to dedup table, idempotent API…