Skip to main content

Introduction to Azure Virtual Machines


Introduction to Azure Virtual Machines



In Microsoft Data-centers, Azure Virtual Machines (VMs) are hosted on Windows Server 2012 R2 Hyper-V servers. But these VMs are different from On-Premises VMs in the following aspects:
  • Console access for Azure VMs can be given by enabling monitoring option.
  • Azure supports Generation 1 VMs only.
  • Azure VMs do not support VHDX format.
  • Azure VMs do not support OS upgrade.
  • Azure VMs depend on the VM size and support more than 1 NICs.
Deployment Scenarios
Typical deployment scenarios of VMs are listed below,
  • Create Test and Dev Environments
  • Extending your data center to the cloud
  • Hosting certain applications in the cloud that can leverage benefits of the cloud. Example - Seasonal Applications.
  • Installing Recovery Site using IaaS-based DRS approach that provides significant costs savings.
  • High-performance computing to solve complex problems involving millions of variables or calculations such as an earthquake.
  • Big data analysis that involves processing and mining massive datasets.
Planning Considerations
While planning for the virtual machine deployment, you have to consider the following,
1.    Suitable and Unsuitable Workloads
2.    Supported and Unsupported Server Roles
3.    Supported and Unsupported Server Features
1. Suitable Workloads
Azure VMs can be used for the following Workloads.
  • Highly available service workloads. E.g., Online Stores.
  • Unpredictable Spikes - E.g., News Channels.
  • Periodic workloads - E.g., Retail sales spurt during Holiday Season.
  • Steady workloads - E.g., Extend or offload existing infrastructure to the cloud.
Unsuitable Workloads
When Planning Azure VMs, it is also important to understand that not every application or service is a suitable fit for the cloud.
Examples
·         Low volume or limited growth workloads - such services or applications can be run on commodity hardware on-premise and will be less expensive than in the cloud.
·         Regulated environment workloads - Certain data is regulated by an organization or the local government. Such restricted and confidential data must be kept on-premises.
2. Unsupported Server Roles
Most of the Windows Roles can be enabled on Azure VMs, but few Roles are not supported in Azure VM.
  • Dynamic Host Configuration Protocol Server
  • Hyper-V (Hyper-V role is supported in Azure Ev3 and Dv3 series VMs only)
  • Rights Management Services
  • Windows Deployment Services
3. Unsupported Server Features
The following significant features are not supported.
  • BitLocker Drive Encryption (on the operating system hard disk, may be used on data disks)
  • Internet Storage Name Server
  • Multipath I/O
  • Network Load Balancing
  • Peer Name Resolution Protocol
  • RRAS
  • DirectAccess
  • SNMP Services
  • Storage Manager for SANs
  • Windows Internet Name Service
  • Wireless LAN Service


Azure Cost Optimization Tools
Now you know the consideration, the next major planning factors with any cloud-based service are:
  • Availability of resources
  • Cost optimization.
To help with estimating the potential costs and to achieve the cost optimization following tools are used:
·         Pricing Calculator tool enables you to estimate the cost of different workloads and services in Microsoft Azure.
·         TCO Calculator estimates the cost savings that can be realized by migrating the application workloads to Microsoft Azure.
Pricing Calculator can be used to estimate the costs for Azure VMs.
VM size and Configuration decides the Cost of the VM
VM sizing is based on,
  • Compute: Capacity required
  • Storage: Size, location, and configuration
  • Disk: Size, persistence and caching
VM configuration deals with,
  • Operating System: Windows Server 2012 R2, Windows Server 2016
  • IP Address allocation: Static or Dynamic IP addresses
  • Availability: Uptime requirements, geo-distribution, service level agreements, and accessibility.
  • Scale set: Type of scaling and threshold

Create Azure VM:





Comments