Plan LLM Capacity as Tokens Per Second on Xeon
Measure prefill and decode tokens per second on Intel Xeon Platinum CPU so a Lucknow VPS queue depth matches real 7B throughput, not marketing.
AI Infrastructure8 min read
NetBay journal
Guides for operators
Practical guides for Linux, containers, CI/CD, self-hosting, and running VPS infrastructure from Netbay Engineering.
499 articles · 24 topics
AllLinuxLinux AdministrationWindowsLinux SecurityLinux NetworkingContainersContainers & OrchestrationKubernetesCI/CDGitHubSelf-HostingBackups & DataDatabasesApp DeploymentDNS & TLSObservabilityEmailAI AgentsAI InfrastructureAPI & AutomationProxiesCloud ArchitectureHardware & PerformanceDevOps Practices
Rolling Updates and Zero-Downtime Deploys in Docker Swarm
Deploy updates to a Swarm service with zero downtime using order start-first, health checks, health thresholds, and automatic rollback.
Containers & Orchestration8 min read
Branch Protection Rules That Actually Protect a Branch
Configure branch protection correctly — required checks, approvals, and stale-review dismissal — so the rule is enforced, not a formality.
CI/CD6 min read
Self-Hosting Uptime Kuma for Status Pages
Monitor your services and publish a public status page with Uptime Kuma, Telegram and webhook alerts, health checks, and a single Docker container.
Self-Hosting6 min read
Idempotent Automation: Why Scripts Must Be Re-Runnable
Write automation that is safe to run twice: idempotent scripts and modules converge to the same state no matter when they are re-run.
API & Automation6 min read
Flaky Checks and Retry Strategies in CI
Diagnose nondeterministic tests, use retries the right way, and build pipelines that are reliable not lucky.
CI/CD7 min read
ReplicaSets and Rolling Updates: How Deployments Self-Heal
See how ReplicaSets keep pod counts stable and how Deployments perform rolling updates with no downtime using readiness probes.
Kubernetes6 min read
Token and Cost Budgets for Long-Running Agents
Cap steps, tokens, and daily spend so a long-running ops agent on a Linux VPS cannot loop itself into an unbounded model-API bill overnight.
AI Agents8 min read
Alpine Linux Beyond Containers: Server Roles and musl Notes
Alpine already dominates containers; as a full server it is a tiny router, DNS, VPN, or monitor. Learn the musl quirks and OpenRC service model here.
Linux6 min read
Set Node max-old-space-size on a 2 GB VPS
Cap the V8 old space on a 2 GB Ubuntu VPS so Node cannot starve the kernel, nginx, and the page cache before the Linux OOM killer fires first.
App Deployment7 min read
Structured JSON Output That Agents Can Parse
Force models to emit JSON Schema, then validate locally before any tool runs, so a stray fence or trailing sentence never crashes your agent loop.
AI Agents8 min read
Understanding Swap, Partitions, and LVM on a VPS
How swap space, disk partitions, and LVM logical volumes work together to manage storage on a Linux VPS.
Linux7 min read