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: azure-stack/operator/azure-site-recovery-capacity-planning.md
+5-5Lines changed: 5 additions & 5 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -4,7 +4,7 @@ description: Learn about capacity planning for Azure Site Recovery.
4
4
author: ronmiab
5
5
ms.author: robess
6
6
ms.topic: conceptual
7
-
ms.date: 08/12/2024
7
+
ms.date: 06/10/2025
8
8
ms.reviewer: rtiberiu
9
9
ms.lastreviewed: 04/15/2024
10
10
---
@@ -79,7 +79,7 @@ The target environment requires one Azure Site Recovery vault to be created for
79
79
Installation of Azure Site Recovery on Azure Stack Hub requires that you install the Site Recovery Resource Provider (RP).
80
80
81
81
> [!NOTE]
82
-
> With Microsoft.SiteRecovery-1.2301.2216.2287, Azure Site Recovery on Azure Stack Hub does not require Event Hubs as a dependency.
82
+
> With Microsoft.SiteRecovery-1.2301.2216.2287, Azure Site Recovery on Azure Stack Hub doesn't require Event Hubs as a dependency.
83
83
84
84
:::image type="content" source="../operator/media/azure-site-recovery/capacity-planning/three-services.png" alt-text="Screenshot of the three services to install Azure Site Recovery on Azure Stack Hub."lightbox="media/azure-site-recovery/capacity-planning/three-services.png":::
85
85
@@ -117,7 +117,7 @@ When creating the BCDR plan, consider all aspects of the protected workloads. Th
117
117
118
118
For the scope of Azure Site Recovery on Azure Stack Hub, here's a starting point for calculations, especially for the cache storage account used:
119
119
120
-
1. If there's a failover, during normal operations, multiply the number of disks replicated by the average RPO. For example, you might have (2MB * 250s). The cache storage account is normally a few KB to 500 MB per disk.
120
+
1. If there's a failover, during normal operations, multiply the number of disks replicated by the average RPO. For example, you might have (2 MB * 250 s). The cache storage account is normally a few KB to 500 MB per disk.
121
121
122
122
2. If there's a failover, given a worst case scenario, multiply the number of disks replicated by the average RPO over a full day.
123
123
@@ -151,11 +151,11 @@ The following table is an example of tests run in our environments. You can use
151
151
|16 |32 MB/s |4096 ||
152
152
153
153
> [!NOTE]
154
-
> 8Kb is the smallest block size of data Azure Site Recovery supports. Any changes less than 8Kb are treated as 8Kb.
154
+
> 8 Kb is the smallest block size of data Azure Site Recovery supports. Any changes less than 8 Kb are treated as 8 Kb.
155
155
156
156
To test further, we generated a consistent type of workload; for example, consistent storage changes in blocks of 8 Kb that total up to 1 MB/s per disk. This scenario isn't likely in a real workload, given that changes can happen at various times of the day, or in spikes of various sizes.
157
157
158
-
To replicate these random patterns, we've also tested scenarios with:
158
+
To replicate these random patterns, we also tested scenarios with:
159
159
160
160
- 120 VMs (80 Windows, 40 Linux) protected through the same Azure Site Recovery VM appliance.
161
161
- Each VM generating at random intervals, at least twice per hour, random blocks totaling 5 Gb of data across five files.
0 commit comments