Microsoft Azure

Cluster Configuration Options

Each Atlas cluster tier comes with a default set of resources. Atlas provides the following resource configuration options:

Custom Storage Size

The size of the server root volume. Atlas clusters deployed onto Azure use premium SSDs. [1]

Note

RAM Availability

The actual amount of RAM available to each cluster tier might be slightly less than the stated amount, due to memory that the kernel reserves.

Note

As of October 18, 2021, the following Atlas clusters deployed to Azure offer 16,000 IOPS (up from 7,500) and 500 MB/second throughput (up from 250 MB/second):

• New clusters with 4 TB storage volumes.

• Existing clusters that you scale up to 4 TB storage volumes.

The following cluster tiers are available:

Cluster Tiers	Storage Range	Default Storage	Default RAM
M0	.5 GB	.5 GB	Shared
M2	2 GB	2 GB	Shared
M5	5 GB	5 GB	Shared
M10	8 GB to 128 GB	8 GB	2 GB
M20	8 GB to 256 GB	16 GB	4 GB
M30	8 GB to 512 GB	32 GB	8 GB
M40	8 GB to 1 TB	64 GB	16 GB
R40	8 GB to 1 TB	128 GB	16 GB
M50	8 GB to 4 TB	128 GB	32 GB
R50	8 GB to 4 TB	128 GB	32 GB
M60	8 GB to 4 TB	128 GB	64 GB
M60_NVME	1600 GB	1600 GB	64 GB
R60	8 GB to 4 TB	128 GB	64 GB
M80	8 GB to 4 TB	256 GB	128 GB
R80	8 GB to 4 TB	256 GB	128 GB
M80_NVME	1600 GB	1600 GB	128 GB
M200	8 GB to 4 TB	256 GB	256 GB
R200	8 GB to 4 TB	256 GB	256 GB
M200_NVME	3100 GB	3100 GB	256 GB
R300	8 GB to 4 TB	512 GB	384 GB
M300_NVME	3600 GB	3600 GB	384 GB
R400	8 GB to 4 TB	512 GB	432 GB
M400_NVME	4000 GB	4000 GB	512 GB
M600_NVME	4000 GB	4000 GB	640 GB

Can use this tier for a multi-cloud cluster.

Not available in the following regions:

• germanywestcentral

• switzerlandnorth

• switzerlandwest

Note

Cluster Tier & API Naming Conventions

For purposes of management with the Atlas Administration API, cluster tier names that are prepended with R instead of an M (R40 for example) run a low-CPU version of the cluster. When creating or modifying a cluster with the API, be sure to specify your desired cluster class by name with the providerSettings.instanceSizeName attribute.

Important

Multi-Cloud Low-CPU clusters

Low-CPU cluster tiers (R40, R50, R60, etc) are available in multi-cloud cluster configurations as long as the cluster tier is available for all the regions that the cluster uses.

Workloads typically require less than 2TB.

NVMe clusters don't support multi-cloud cluster configurations.

Atlas configures the following resources automatically and does not allow user modification:

Encrypted Storage Volumes

Azure storage volumes are always encrypted.

Azure Availability Zones

Azure maintains multiple data centers within each region. Azure groups the data centers into availability zones, which are separate locations within the region. Maintaining data centers in different physical locations helps Azure tolerate local failures.

Azure availability zones aren't available in all regions. To learn which Azure regions maintain availability zones, see the Azure Region table. In regions where availability zones aren't yet available, Azure uses fault domains to help ensure failure tolerance.

Atlas uses Azure availability zones automatically when you deploy a dedicated cluster to a region that supports them. Atlas splits the cluster's nodes across availability zones. For example, a three-node replica set cluster would have one node deployed onto each zone. A local failure in the Azure data center hosting one node doesn't impact the operation of data centers hosting the other nodes.

Azure Fault Domains

Each Azure region includes a set number of fault domains for failure tolerance. Fault domains consist of a group of virtual machines that share a common power source and network switch. If you deploy your cluster to a region that doesn't support availability zones, Atlas spreads the nodes across the fault domains instead.

Atlas uses availability sets to deploy clusters across fault domains. For regions that have at least three fault domains (3FD), Atlas deploys clusters across three fault domains. For regions that only have two fault domains (2FD), Atlas deploys clusters across two fault domains.

The Atlas Add New Cluster form marks regions that support 3FD clusters as Recommended, as they provide higher availability.

The number of fault domains in a region has no effect on the number of MongoDB nodes Atlas can deploy. MongoDB Atlas clusters are always made of replica sets with a minimum of three MongoDB nodes.

For general information on Azure fault domains and availability sets, see Availability Sets Overview

Regions with at Least Three Fault Domains

If the selected Azure region has at least three fault domains, Atlas clusters are split across three fault domains. For example, a three node replica set cluster would have one node deployed onto each zone.

click to enlarge

3FD clusters have higher availability compared to 2FD clusters. However, not all regions support 3FD clusters.

Regions with Only Two Fault Domains

If the selected Azure region has two fault domains, Atlas clusters are split across the two fault domains. For example, a three node replica set cluster would have two nodes deployed to one zone and the remaining node deployed to the other zone.

click to enlarge

2FD clusters have a higher chance of loss of availability in the event of the loss of an zone than 3FD clusters. However, where latency or location are a priority, a region that supports 2FD clusters may be preferred.

[1]	For detailed documentation on Azure storage options, see High-performance Premium Storage and managed disks for VMs

Dedicated Search Nodes

Integrations

Along with global region support, the following product integrations enable applications running on Azure, such as Azure Virtual Machines, Azure Functions, and Azure Container Instances, to use Atlas instances easily and securely.

More Information

For more information on how to use Azure with Atlas most effectively, review the following best practices, guides, and case studies:

• Power BI Desktop: Connect to Atlas from Power BI Desktop

• Visual Studio (VS) Code: Work with MongoDB from VS Code