Tag: performance audit

Abstract grid of glowing compute cells densely packed into reserved cluster capacity

Spark on Kubernetes Reserves CPU It Never Uses. Here’s the Overcommit Fix.

Spark sets executor CPU requests equal to limits, so a Kubernetes cluster reserves twice the CPU it uses and refuses to schedule pending pods. Kubernetes has no native overcommit. Here is the mutating-webhook operator I use to fix it.

By jlu, 2 weeks ago

data

Stop using MinIO as a NoSQL database — why S3 object stores collapse on small-file workloads

Two MinIO platforms, same root cause: used as a NoSQL store. Field notes on LIST IOPS, XFS directory limits, scanner & heal SLAs, the erasure-coding storage-efficiency inversion on small objects, and why Apache Cassandra (or Ceph) is the right answer on-prem in 2026.

By jlu, 4 weeks2026-05-17 ago