COMPUTE

Right-sizing compute: cutting cloud spend without downtime

Most clusters are paying for headroom they never use. We trimmed compute spend substantially while improving tail latency — here is the method.

Right-sizing compute: cutting cloud spend without downtime

Over-provisioning feels safe and quietly doubles the bill. The fix is to scale on the signal that actually predicts load.

Scale on the right signal

CPU is a lagging indicator. We autoscale on request concurrency and queue depth, so capacity arrives before latency suffers.

  • Concurrency-based autoscaling instead of raw CPU
  • Right-sized requests and limits informed by real percentiles
  • Bin-packing workloads to reclaim stranded capacity

Prove it with percentiles

Every change is validated against p95 and p99 latency, not averages — cheaper should never mean slower for the unlucky request.

Build on the same network

Deploy to 30 regions on the anycast edge — free to start.

Start building free
Stay in the loop

Engineering deep-dives, monthly

Join 40,000 builders getting our best writing on infrastructure, performance, and reliability. No spam, unsubscribe anytime.