📚 Series

Azure - Deploy and Manage Compute Resources

...

Episodes (4)

1 Sequence 1 — The Azure Compute Landscape: Choosing the Right Service Azure's compute catalog looks crowded, but it follows one axis: how much control you keep versus how much management you hand to Azure. This lesson decodes the IaaS–PaaS–serverless spectrum and the shared responsibility model, introduces all seven main compute services (VMs, Scale Sets, AKS, ACI, Container Apps, App Service, Functions), and gives you four decision criteria plus a decision tree to place any workload. It also lays the foundations reused across the whole module: regions, availability zones, and the resource hierarchy from resource to management group. Directly targets the AZ-104 'Deploy and manage Azure compute resources' objective — the 'where should this workload run?' question that opens most exam scenarios.
2 Sequence 2 — Virtual Machines: Anatomy and Provisioning A VM is never just 'a VM': every deployment creates a graph of resources — OS disk, NIC, public IP, NSG — each with its own lifecycle, cost, and security implications. This lesson dissects that anatomy, teaches you to decode any VM size name (family letter, vCPU count, feature letters, generation), and walks through image sources (Marketplace, custom, Azure Compute Gallery) including the generalized-versus-specialized distinction. You provision with az vm create, learn what its convenience defaults silently create, bootstrap with cloud-init versus extensions and Run Command, and master the classic billing trap: Stopped still bills compute, only Stopped (deallocated) doesn't. Core AZ-104 question territory.
3 Sequence 3 — VM Disks, Security, and Day-2 Administration Disks are where VMs waste money or lose data: this lesson dissects the OS, temp, and data disk roles (and why the ephemeral temp disk must never hold real data), then walks the five managed disk types with the criteria that separate them — including the rule that Premium SSD v2 and Ultra are data-disks-only and the Premium-SSD condition on the single-VM 99.9% SLA. You cover everyday operations (expand-only resizing, incremental snapshots, host caching per disk role), untangle the three encryption options with the 'Rest, Guest, Best' hook (SSE PMK/CMK, ADE, encryption at host), remove public RDP/SSH exposure with Azure Bastion and just-in-time access, and finish with the day-2 toolkit: resize, redeploy, boot diagnostics, and Azure Update Manager. Classic AZ-104 question territory throughout.
4 Sequence 4 — High Availability and Scaling: Sets, Zones, and Scale Sets A single VM, however well managed, is a single point of failure — this lesson makes compute survive failures and grow with demand. You name the threat first (update domains for planned maintenance, fault domains for rack failure, availability zones for datacenter loss), climb the financially backed SLA ladder (99.9% single VM on Premium disks, 99.95% availability set, 99.99% across zones), then add elasticity with VM Scale Sets: Flexible versus Uniform orchestration, autoscale anchors (min/max/default), paired metric rules with a 70/30 dead band to avoid flapping, schedule-based profiles, and scale-in policy. Closes with vertical versus horizontal scaling and the rule that scale-out always sits behind a load balancer. Guaranteed AZ-104 exam territory.