You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: articles/virtual-machines/disks-high-availability.md
+6-4Lines changed: 6 additions & 4 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -3,7 +3,7 @@ title: Best practices for high availability with Azure VMs and managed disks
3
3
description: Learn the steps you can take to get the best availability with your Azure virtual machines and managed disks.
4
4
author: roygara
5
5
ms.author: rogarana
6
-
ms.date: 05/07/2024
6
+
ms.date: 05/09/2024
7
7
ms.topic: conceptual
8
8
ms.service: azure-disk-storage
9
9
---
@@ -29,6 +29,8 @@ Before going over recommendations for achieving higher availability, you should
29
29
30
30
Managed disks are designed for 99.999% availability and provide at least 99.999999999% (11 9’s) of durability. With managed disks, your data is replicated three times. If one of the three copies becomes unavailable, Azure automatically spawns a new copy of the data in the background. This ensures the persistence of your data and high fault tolerance.
31
31
32
+
:::image type="content" source="media/disks-high-availability/disks-lrs-zrs-diagram.png" alt-text="Diagram of LRS and ZRS." lightbox="media/disks-high-availability/disks-lrs-zrs-diagram.png":::
33
+
32
34
Locally redundant storage (LRS) disks provide at least 99.999999999% (11 9's) of durability over a given year and zone-redundant storage (ZRS) disks provide at least 99.9999999999% (12 9's) of durability over a given year. This architecture helps Azure consistently deliver enterprise-grade durability for infrastructure as a service (IaaS) disks, with an industry-leading zero percent [annualized failure rate](https://en.wikipedia.org/wiki/Annualized_failure_rate).
33
35
34
36
## Recommendations for applications running on a single VM
@@ -72,7 +74,7 @@ With zone-redundant Virtual Machine Scale Sets using the flexible orchestration
72
74
73
75
#### Deploy VMs and disks across three availability zones
74
76
75
-
Another method to spread VMs and disks across availability zones is to manually deploy the VMs and disks across three availability zones. This deployment provides redundancy in VMs and disks across multiple data centers in a region, allowing you to fail over to another zone if there's a data center or zonal outage.
77
+
Another method to spread VMs and disks across availability zones is to deploy the VMs and disks across three availability zones. This deployment provides redundancy in VMs and disks across multiple data centers in a region, allowing you to fail over to another zone if there's a data center or zonal outage.
76
78
77
79
78
80
### Deploy VMs and disks across multiple fault domains
@@ -88,6 +90,8 @@ To deploy resources across multiple fault domains, you can either use [regional
88
90
89
91
#### Use regional Virtual Machine Scale Sets with flexible orchestration
90
92
93
+
:::image type="content" source="media/disks-high-availability/disks-availability-set.png" alt-text="Diagram of availability set." lightbox="media/disks-high-availability/disks-availability-set.png":::
94
+
91
95
A regional Virtual Machine Scale Set is a Virtual Machine Scale Set that has no explicitly defined availability zones. With regional virtual machine scale sets, VM resources are replicated across fault domains within the region they're deployed in to improve the resiliency and availability of applications and data. This configuration spreads VMs across fault domains by default but also provides the ability to assign fault domains on VM creation. See [this section](../virtual-machine-scale-sets/virtual-machine-scale-sets-use-availability-zones.md#regional) for details.
92
96
93
97
Regional Virtual Machine Scale Sets don't protect against large-scale outages like a data center or region outage, and don't currently support Ultra Disks or Premium SSD v2 disks.
@@ -96,8 +100,6 @@ Regional Virtual Machine Scale Sets don't protect against large-scale outages li
96
100
97
101
[Availability sets](availability-set-overview.md) are logical groupings of VMs that place VMs in different fault domains to limit the chance of correlated failures bringing related VMs down at the same time. Availability sets also have better VM to VM latencies compared to availability zones.
98
102
99
-
:::image type="content" source="media/disks-high-availability/disks-availability-set.png" alt-text="Diagram of availability set." lightbox="media/disks-high-availability/disks-availability-set.png":::
100
-
101
103
Availability sets don't let you select the fault domains for your VMs, can't be used with availability zones, don't protect against data center or region-wide outages, and don't currently support Ultra Disks or Premium SSD v2 disks.
0 commit comments