Edit

Disks and cost optimization

  • Article

Azure managed disks are block-level storage volumes that are managed by Azure and used with Azure Virtual Machines. Managed disks are like a physical disk in an on-premises server, but these disks are virtualized. Azure supports many types of disks. For more information, reference Azure managed disk types.

To understand how Azure managed disks are cost-effective solutions for your workload, reference the following articles:

The following sections include design considerations, a configuration checklist, and recommended configuration options specific to Azure managed disks and cost optimization.

Design considerations

Azure Disks include the following design considerations:

  • Use a shared disk for workload, such as SQL server failover cluster instance (FCI), file server for general use (IW workload), and SAP ASCS/SCS.
  • Consider selective disk backup and restore for Azure VMs.
  • Premium storage also features free bursting, combined with an understanding of workload patterns, offers an effective SKU selection and cost optimization strategy for IaaS infrastructure, enabling high performance without excessive over-provisioning and minimizing the cost of unused capacity.
Considerations Description
Use a shared disk for workload, such as SQL server failover cluster instance (FCI), file server for general use (IW workload), and SAP ASCS/SCS. You can use shared disks to enable cost-effective clustering instead of setting up your own shared disks through S2D (Storage Spaces Direct). Sample workloads that would benefit from shared disks include:
- SQL Server Failover Cluster Instances (FCI)
- Scale-out File Server (SoFS)
- File Server for General Use (IW workload)
- SAP ASCS/SCS

Checklist

Have you configured your Azure managed disk with cost optimization in mind?

  • Configure data and log files on different disks for database workloads.
  • Use bursting for P20 and lower disks for workloads, such as batch jobs, workloads, which handle traffic spikes, and to improve OS boot time.
  • Consider using Premium disks (P30 and greater).

Configuration recommendations

Consider the following recommendations to optimize costs when configuring your Azure managed disk:

Recommendation Description
Configure data and log files on different disks for database workloads. You can optimize IaaS DB workload performance by configuring system, data, and log files to be on different disk SKUs (leveraging Premium Disks for data and Ultra Disks for logs satisfies most production scenarios). Ultra Disk cost and performance can be optimized by taking advantage of configuring capacity, IOPS, and throughput independently. Also, you can dynamically configure these attributes. Example workloads include:
- SQL on IaaS
- Cassandra DB
- Maria DB
- MySql and
- Mongo DB on IaaS
Use bursting for P20 and lower disks for workloads, such as batch jobs, workloads, which handle traffic spikes, and to improve OS boot time. Azure Disks offer various SKUs and sizes to satisfy different workload requirements. Some of the more recent features could help further optimize cost performance of existing disk use cases. You can use disk bursting for Premium (disks P20 and lower). Example scenarios that could benefit from this feature include:
- Improving OS boot time
- Handling batch jobs
- Handling traffic spikes
Consider using Premium disks (P30 and greater). Premium Disks (P30 and greater) can be reserved (one or three years) at a discounted price.
Optimize with managed disks. Determine your performance needs in combination with your storage capacity needs, accounting for fluctuating workload patterns. Knowing your needs allows you to determine what disk type and disk size you need. Some higher performance disk types offer extra cost optimization features and strategies.
Consider Ephemeral OS disks. Ephemeral OS disks provide top-tier performance at no extra cost, but are non-persistent, have limited capacity, and are restricted to OS and temp disk use only.

Next step

Event Grid and reliability