Infrastrukturní architektura — Knowledge Base
Přehled témat, principů a best practices pro návrh a provoz infrastruktury.
Obsah
| Oblast | Soubor | Popis |
|---|---|---|
| ☁️ Cloud architektura | CLOUD.md | AWS/Azure/GCP, hybrid cloud, multi-cloud, well-architected framework |
| 🌐 Síťová architektura | NETWORKING.md | DNS, load balancing, firewall, VPN, CDN, BGP, TCP/IP |
| 📊 Monitoring a observabilita | MONITORING.md | Logging, metrics, tracing, alerting, SLI/SLO/SLA |
| 🔄 CI/CD a DevOps | CICD.md | Pipelines, GitOps, IaC (Terraform, Pulumi), konfigurace |
| 🗄️ Databázová architektura | DATABASES.md | SQL/NOSQL, sharding, replikace, caching, migrace |
| 🖥️ Hypervisory | HYPERVISORS.md | VMware, Hyper-V, KVM, Proxmox, virtualizační platformy |
| 🏭 Datová centra | DATACENTERS.md | Tier klasifikace, power, cooling, Aisle containment, ASHRAE |
| 💾 Storage | STORAGE.md | SAN/NAS/DAS/object, RAID, SDS, Ceph |
| 🔧 Server hardware | SERVER-HW.md | CPU (Xeon/EPYC), RAM, PCIe, NUMA, TDP, BMC, storage controllers |
| 🔌 Server connectivity | CONNECTIVITY.md | Ethernet (1-800 GbE), FC SAN, iSCSI, NVMe-oF, SAS, NIC features |
| 🎮 GPU | GPU.md | NVIDIA/AMD GPU, NVLink, MIG/vGPU, AI training/inference |
| ⚙️ Server config | SERVER-CONFIG.md | BIOS tuning, DB/hypervisor/K8s/storage best practices |
| 📦 Provisioning | PROVISIONING.md | PXE/iPXE, Kickstart, Redfish, Terraform, config management |
| 📋 Původní HARDWARE | HARDWARE.md | Rozděleno na SERVER-HW, GPU, SERVER-CONFIG, PROVISIONING |
| 📋 Review workflow | REVIEW.md | Proces oponentury a kontroly obsahu |
| 📝 ADR template | templates/ADR.md | Architecture Decision Record template |
Zdroje
Raw referenční data (dokumentace, knihy, standardy) podle oblastí:
- sources/cloud/, sources/networking/, sources/monitoring/
- sources/cicd/, sources/databases/, sources/infrastructure/
Principy
- Dostupnost — SLA, redundance, failover, multi-AZ/region
- Škálovatelnost — horizontalní vs. vertikální, auto-scaling
- Bezpečnost — defense in depth, least privilege, zero trust
- Náklady — FinOps, right-sizing, reserved instances
- Operability — observabilita, automation, dokumentace