A Lap around Azure Compute Services

Microsoft Azure Cloud comprises of around nearly five hundred services (as of this writing) across versatile categories like Compute, Storage, Network, Database, Security, DevOps, Identity, Analytics, Containers, AI, IoT , Machine Learning, AR/VR etc.

In this blog post, will be sharing a quick glimpse of each Azure compute services available across its categories as of writing date.

Azure Compute services offering

Compute: The Compute category provides resources around virtualization and computation across dedicated hosts, batch based services, server based and server less migration of lift-shift of application deployment.

— The Compute services are further classified as IaaS, PaaS, Serverless and Microservices, high performance computing (HPC) workloads and Hybrid Cloud workloads

-Infrastructure as a Service (IaaS)

• Virtual Machines — on demand, scalable computing resources Azure offers.
• Virtual Machine (classic) — VMs deployed through other than Azure resource manager deployment.
• Virtual Machine Scale Sets — VM scale sets can help to deploy load balanced identical set of unified linux / windows VM instances based on autoscale groups and specific scale out and scale in rules.
• Availability Sets — Logical grouping of Azure virtual machines deployed across the multiple servers, compute racks, storage units and network switches within the same DC. At least two instances of VM deployed on an availability set provides 99.95% of SLA. VMs part of same fault domain within the DC shares the same compute power source & network switch.
• Images — This refers to standard OS images applicable for Azure VMs from marketplace based on Windows & Linux offerings. This images are offered for both Gen1 & Gen2 VMs which can be stored on Azure Blob storage. The VM images also provide capabilities like Zonal resiliency snapshots which means the image snapshots are stored across the availability zones of a specific region.
• Image Templates — Since Azure Image builder simplifies the image customization pipeline, hence a configuration setting for the images can be defined using Image templates. The source images can be imported from Azure marketplace, Azure Managed images, Azure compute gallery with distribution targets like VM image gallery, managed image and storage blob. Azure Image Builder also allows for validation of these both platform & customized images with the parameters like “Continue to distribution on failure” which allows the output target images will be distributed even if a failure is encountered.
• VM Image definitions — VM image definitions are defined within an Azure Compute Gallery and carry information about the image and requirements for using it. A VM image definition may have one or more VM image versions. There are three parameters for each image definition.

-Image Definition

-Publisher

-Offer

-Sku

• VM Image Versions — A VM image version can be used to create a VM when using an Azure compute gallery. You can create a VM in any region where the VM image version is replicated. The source images can be imported from disks or snapshots, VM Image version, managed image or storage blobs (VHD). A VM image version can be used to create a VM when using an Azure compute gallery. You can create a VM in any region where the VM image version is replicated. The VM image versions can be encrypted using platform-managed or customer-managed key.
• Community Images — The community image refers to standardized custom images available from Azure Image gallery. These public images are available from different public galleries like ‘aquasecdev’, ‘cloudyspells’ etc. Images can be agentless or agent based.
• VM Application Definitions — VM app definitions are community images which allows simplifying management, sharing and global distribution of apps within the VMs.
• VM Application Versions — Refers to deployable resource versions differs by app versions, OS versions, link to app package file from the storage account, install string while installing the app, package file name, configuration file name, update string name, target region for replication, replica count per replication etc.
• Host Group — collection of dedicated hosts which can be created in an Azure region spread across the availability zones. It specifies the required fault & update domain counts enables the deployment of Ultra SSD disks for I/O intensive workloads.
• Disk pools — collection of dedicated disks which can be provisioned at Azure region with the specific availability zone. In that scenario, the managed disk and the public IP (if any) will be deployed on the same AZ. The Sku tier (standard SSD, premium SSD)can be added to the disk pool which are deployed at the same Azure region and availability zone. iSCSI protocol can be integrated with the disk pool and one iSCSI target can be configured and added to disk pool.
• Proximity Placement Group — proximity placement group is a logical grouping allows compute resources to be deployed at close location inside the datacenter while maximizing throughput and minimizing latency. Proximity Placement Groups can be configured for Azure VMs starting from B series, D series, E Series & M series.
• Restore Point Collection — VM Restore Point collections consist of restore points of each specific VM and VM restore point contains the disk restore point attached to the VM. A resource point stores virtual machine configurations and point-in-time crash (if the VM is shutdown) or application consistent snapshot for all managed disks attached to a virtual machine. Both of OS & data disks can be restored to a particular restore point integrated with date and time.
• Azure Virtual Desktop — AVD provides virtualized desktop services with Windows 10 and Windows 11 OS images allowing seamless user experience deployed over dedicated host pools and configured through application host groups.

#Happy Cloud Journey with Azure