Tag
4 articles tagged #Reliability.
P.01Use a monotonic clock for durations and timeouts, the wall clock only for timestamps. Python, JavaScript and Go code, plus the leap-second bug that shows why.
P.02Your p99 is slower than your average suggests, and fan-out makes it worse. How hedged requests and load shedding cut tail latency, with TypeScript code.
P.03A write-ahead log makes a database durable by recording changes before touching data files. How WAL commits, recovers, replicates, and what it trades.
P.04Exponential backoff plus jitter stops retries from turning a blip into an outage. Working code, what not to retry, and how to bound the loop.