How We Found 7 TiB of Memory Just Sitting Around

October 30, 2025Render BlogView original ↗

How we saved 7 TiB of memory across our Kubernetes clusters by disabling namespace listwatching in Vector, reducing daemonset overhead and API server load at scale.

From other products

Workers, Durable Objects - Track memory usage for Workers and Durable Objects in the dashboardJun 30, 2026
Memory usage charts now in Workers Metrics tab
New Memory Usage chart in the Workers Metrics tab shows P50, P90, P99, and P999 memory percentiles for Workers and Durable Objects, with de…
Cloudflare · Developer Platform
How we saved over $3 million in idle compute costs with Datadog Kubernetes AutoscalingJun 25, 2026
$3M saved through Kubernetes Autoscaling internal adoption
Datadog's internal Rapid team cut idle compute costs by more than 50% in its first data center rollout by adopting Datadog Kubernetes Autos…
Datadog · Datadog Blog

Fetched April 11, 2026

How We Found 7 TiB of Memory Just Sitting Around — Render… — releases.sh

How We Found 7 TiB of Memory Just Sitting Around

More from Render

From other products

More from Render

From other products