News
Uber Cuts Translation Cache Memory by up to 90% While Keeping P99 Latency Below 1 MS
InfoQ · By Leela Kumili ·
The Brief
- Uber redesigned its Go-based translation cache to cut its initial memory footprint by more than 90%.
- The cache keeps a hit rate above 99.9% and p99 disk-backed cache miss latency below 100 microseconds.
- The hybrid architecture keeps in-memory keys and frequently accessed data, with the rest backed by disk.
Why it matters
Teams running large in-memory caches can cut memory costs substantially with a hybrid memory and disk design while keeping tail latency low.
Who’s affected
- Engineers running Go-based caching layers at scale
Uber redesigned its Go-based translation cache to reduce its initial memory footprint by more than 90% while maintaining a cache hit rate above 99.9% and p99 disk-backed cache miss latency below 100 microseconds. The hybrid architecture combines in-memory keys and frequently…