Introduction
We all know the concept of IoT and use it in different devices to connect them virtually. So in this blog, you'll learn the Azure IoT Hub concept. The cloud-hosted Azure IoT Hub managed service acts as the primary message hub for communications between IoT applications and linked devices. Connecting millions of devices safely and dependably with their backend services is possible. Almost every device can be connected to an IoT hub.
Let's dive in more about Scalability and Availability in Azure IoT Hub.
Availability Zones
IoT Hub supports Availability Zones. A high-availability product called an availability zone shields your apps and data from data center outages. Each zone offers one or more data centers with independent power, cooling, and networking at a distinct physical location. Within the region, this provides replication and redundancy.
- East Australia
- South Brazil
- Central Canada
- Central US
- Central France
- West Central Germany
- East Japan
- North Europe
- Southeast Asia
- South UK
- West US 2
The broad areas we are going to discuss in this article are given below:
- Intra-region HA
- Cross-region DR
- Achieving cross-region HA
Let's learn more about this one by one. First, let's have a look at
Intra-region HA
The IoT Hub service implements redundancy in nearly all of the service's levels to offer intra-region HA. To utilize these HA features, IoT solution developers don't need to make any more effort. Despite its reassuringly high uptime guarantee, connection failures are still possible with IoT Hub, just like with any distributed computing platform. Moving your solutions from an on-premises solution to the cloud requires a change in emphasis from optimizing "mean time between the failures" to "mean time to recover." In other words, temporary failures must be accepted as usual while operating with the cloud in the mix.
Cross-region DR
There may be some uncommon circumstances where a data center encounters protracted outages due to power outages or other physical asset issues. The center region HA capacity detailed above may not always be helpful in such uncommon circumstances. IoT Hub offers a variety of recovery options after such protracted outages.
Customers in this situation have two recovery options:
Manual failover
The IoT Hub service includes a manual failover feature that lets users switch their hub's activities from one primary region to the equivalent Azure geo-paired region. Manual failover can be performed if there is a local catastrophe or a protracted service outage.
Microsoft-initiate failover
Microsoft occasionally uses "Microsoft-initiated failover" to failover every IoT hub in an impacted region to the matching geo-paired region. This procedure doesn't need the user's input as a default setting. When this option is used is something Microsoft maintains the right to decide. This technique doesn't ask for user permission before taking over the user's hub. The recovery time objective (RTO) for Microsoft-initiated failover is 2-26 hours.
The difference is that Microsoft starts the former, while the user starts the latter.
Achieving cross-region HA
Suppose the RTO provided by either Microsoft-initiated failover or manual failover alternatives does not satisfy your company's uptime objectives. In that case, you should think about creating a per-device automatic cross-region failover method.
The system's back end operates predominantly in one data center located in a regional failover approach. A backup IoT hub and back end are set up in another data center. Devices use a secondary service endpoint if the primary IoT hub in the primary area is down or there is a disruption in network connectivity from the device to the primary region. Instead of sticking to one region, you can increase the solution's availability by using a cross-region failover strategy.
19+ registered 
