PepiteData Julien Laurenceau
  • Home
  • Services
  • Audits
  • Expert Call
  • About
  • Blog
  • Contact

Kubernetes

One gold object fanning out into six copies arranged as two groups of three, one per zone
ceph

Ceph in Production: RBD, RGW and CephFS

A field guide to Ceph in production: which surface serves which workload, what failure domains cost in replicas, how the upgrade lifecycle really rolls back, and what key rotation does not do.

By Julien Laurenceau, 2 months2026-09-30 ago
Cartoon of an oversized GPU processing machine waiting for a single data block from a narrow feed while its billing meter keeps turning.
AI

GPU Cost Optimization: Find the Waste Before Buying More

GPU cost optimization starts with useful work, latency and paid capacity. A measurement-led review of data stalls, inference sizing, sharing and cost attribution separates removable spend from capacity gains.

By Julien Laurenceau, 5 months2026-06-14 ago
Abstract grid of glowing compute cells densely packed into reserved cluster capacity
apache spark

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 Julien Laurenceau, 6 months2026-05-31 ago
apache spark

Apache Spark smoke tests on kubernetes

Introduction Benchmarking remains a critical (and often underestimated) tool when designing or validating large-scale data platforms. While many teams rely on synthetic workloads or production replays, standardized benchmarks still play a key role when comparing architectures, tuning clusters, or validating infrastructure choices. TPCx-HS is one of those benchmarks: designed to Read more

By Julien Laurenceau, 10 months2026-01-12 ago
data

Who’s Using Ceph RBD Mirroring for Kubernetes Storage in Production?

RBD mirroring in production is one of the patterns I audit and design: Ceph storage consulting. Alpha Feature in Production… Good Idea? The most promising approach, journal-based mirroring, offers near real-time replication and faster failover. However, it’s currently an alpha feature in the Ceph CSI driver and relies on rbd-nbd, Read more

By Julien Laurenceau, 11 months2025-12-15 ago
Three open hard drives side by side, the third one on fire: the failure every Kubernetes PVC backend has to plan for
ceph

Choosing the Right Storage Backend for Kubernetes PVCs

Four open source Kubernetes PVC backends compared: Ceph RBD, OpenEBS Local PV ZFS, Mayastor and DRBD with LINSTOR. Whether the application already replicates decides the choice; Ceph RBD brings snapshots and backups, but its multi-site mirroring is not production-ready, unlike Ceph RGW multi-site.

By Julien Laurenceau, 11 months2025-12-15 ago
  • Privacy Policy
  • Mentions légales
  • CGV
  • Cookies
Hestia | Developed by ThemeIsle