Writing
Distributed Systems
Consistency, availability, failure modes, and the hard problems of building systems that span machines.
Distributed Locking: Redis vs etcd
Redis Redlock is fast and familiar. etcd is correct. The choice depends on whether you can tolerate the failure modes of each.
Distributed SystemsRedisetcd
April 1, 2024
8 min read
Backpressure Strategies in High-Traffic APIs
Without backpressure, a traffic spike doesn't just slow your system down — it takes it down. Here are the patterns that actually work.
Distributed SystemsGoReliability
September 18, 2023
7 min read