Skip to main content

226 posts tagged with "Cost Optimization"

Cloud vs owned-hardware cost breakdowns and bill analysis

View all tags

Read the Hosting cost optimization guide

Hetzner's 45-Hour ASH Load Balancer Outage: What a Managed LB Costs a Fleet That Doesn't Run Its Own
·Dora Noda·9 min

Hetzner's 45-Hour ASH Load Balancer Outage: What a Managed LB Costs a Fleet That Doesn't Run Its Own

A Hetzner-managed Load Balancer in ASH went down for 45 hours on July 13, 2026, taking every healthy tenant behind it offline. Here's the actual blast-radius math, and what running kube-vip or MetalLB yourself instead would have cost and bought back.

self-hosting
PaaS
Kubernetes
infrastructure
+1
Kubernetes 1.36 Ships the Metric That Finally Proves Your Route Controller Isn't Wasting API Calls
·Dora Noda·9 min

Kubernetes 1.36 Ships the Metric That Finally Proves Your Route Controller Isn't Wasting API Calls

Kubernetes 1.36 adds a counter that proves whether your route controller is syncing on real node changes or wasting API calls on a timer. Here's what it measures, how to check if your Hetzner CCM exposes it, and a PromQL alert that catches a stuck node join before it becomes an outage.

Kubernetes
infrastructure
self-hosting
PaaS
+1
Kubernetes 1.36's HPA Can Now Scale to Zero: What It Actually Changes for Idle-App Billing
·Dora Noda·9 min

Kubernetes 1.36's HPA Can Now Scale to Zero: What It Actually Changes for Idle-App Billing

Kubernetes 1.36 finally defaults HPAScaleToZero on, but it only solves half the scale-to-zero problem. Here's the exact split, real cold-start numbers from Render and Fly.io, and a worked cost model for a self-hosted PaaS's dormant-app billing.

Kubernetes
PaaS
self-hosting
cost-optimization
+1
What Self-Hosting OpenMeter's Kafka+ClickHouse Stack Actually Costs a Git-Push PaaS
·Dora Noda·8 min

What Self-Hosting OpenMeter's Kafka+ClickHouse Stack Actually Costs a Git-Push PaaS

OpenMeter's self-hosted stack adds two new stateful services — Kafka and ClickHouse — to bill per millisecond. Here's what that actually costs in Hetzner hardware and ops burden against billing tenants on signals a PaaS already tracks for free.

self-hosting
PaaS
infrastructure
engineering
+1
Railway's Three Years Without a Price Increase vs Render's April 2026 Repricing: What Sticker Stability Actually Signals
·Dora Noda·8 min

Railway's Three Years Without a Price Increase vs Render's April 2026 Repricing: What Sticker Stability Actually Signals

Railway hasn't touched its plan fee or rate card since 2023; Render restructured its entire workspace pricing in April 2026. Here's what the two timelines actually reveal about cost structure, growth strategy, and why a flat sticker price isn't the same claim as a flat bill.

self-hosting
PaaS
cost-optimization
infrastructure
Zeabur's $5 'No Sleep' Credit vs a €4.35 Hetzner Box: What You're Actually Paying For
·Dora Noda·9 min

Zeabur's $5 'No Sleep' Credit vs a €4.35 Hetzner Box: What You're Actually Paying For

Zeabur's $5/month Dev plan sells freedom from cold-start sleep. A line-by-line recompute of what that $5 credit actually covers on Zeabur's metered RAM pricing versus a €4.35 Hetzner CX22 that was never billed by the minute at all.

self-hosting
PaaS
cost-optimization
infrastructure
Blacksmith and Depot Colocate Docker Layer Cache on NVMe: What 40x Faster CI Builds Mean for Self-Hosted Build Pipelines
·Dora Noda·9 min

Blacksmith and Depot Colocate Docker Layer Cache on NVMe: What 40x Faster CI Builds Mean for Self-Hosted Build Pipelines

Blacksmith and Depot claim 20x-40x faster Docker builds by moving the layer cache onto NVMe disk colocated with the build machine. Here's the actual mechanism, and what it means for a self-hosted PaaS that already owns its build hardware.

self-hosting
PaaS
Kubernetes
cost-optimization
+1
Dokploy vs. Coolify in 2026: What Happens the Day You Click 'Add Server'
·Dora Noda·9 min

Dokploy vs. Coolify in 2026: What Happens the Day You Click 'Add Server'

Coolify and Dokploy both added multi-server features in 2026. Neither provisions a machine, heals a dead node, or wires up a load balancer without a human at the keyboard. Here's the feature-by-feature gap against a Cluster API fleet.

self-hosting
PaaS
Kubernetes
infrastructure
+1
eBPF Made Continuous Profiling Nearly Free — But Someone Still Has to Store It: Parca vs. Pyroscope on a Self-Hosted Fleet
·Dora Noda·8 min

eBPF Made Continuous Profiling Nearly Free — But Someone Still Has to Store It: Parca vs. Pyroscope on a Self-Hosted Fleet

Parca and Pyroscope both run eBPF-based continuous profiling under 1% CPU overhead. Here's what standing either one up as a Cluster API service actually costs in storage and operational surface, worked against Datadog and Grafana Cloud's own profiling bills for the same fleet.

self-hosting
PaaS
Kubernetes
engineering
+1
Showing 109–117 of 226 posts
Prev13 / 26Next