![]() Ensure you understand the business requirements for your solution. The choices you make for device hardware, connectivity and protocols, and cloud services affect the reliability needs of your solution.Īs described in the Azure WAF – Reliability pillar, desired reliability is subjective. ![]() As described in the Azure WAF – Reliability pillar, the cloud services in your IoT solution must implement reliability principles to provide high availability for your overall IoT solution.īuilding a reliable IoT solution requires careful consideration for devices, cloud, and how they interact. IoT solutions must provide reliability for the cloud services required for the IoT devices to integrate and transmit data. The cloud is often the final data ingress point for all messages flowing from the devices. Networks may have intermittent connectivity issues and in this case the device needs to manage its behavior accordingly.ĭevice management and modeling layer: The cloud service provides each device with an identity and manages your devices at scale. Depending on the type of IoT solution, connectivity reliability is in your hands or the hands of the network service provider. Transport layer: To connect to the cloud service for data, control, and management, devices need access to a network. The reliability of the application running on a device that has supervisory control is crucial. Actions send data back to the machines or environment that the device is authorized to perform supervisory actions on. The solution design should ensure reliable device management and reliable communications from the device to the cloud.ĭevices that provide supervisory control not only collect data to send to the cloud, but also take actions based in that data. Typically, devices and gateways perform the following functions:ĭata collection means the device is connected to sensors or subscribes to telemetry from downstream systems and then pushes this data to the cloud. To achieve overall solution reliability, each layer should have acceptable levels of reliability.ĭevice and gateway layer: Devices and gateways come in many forms and shape. Adding processes for monitoring and notification beyond what's provided by the cloud operations teams.ĭesigning for reliability in an IoT solution should take into consideration the foundational layers in the IoT architecture.Building redundancy into the workload-specific architecture.Making architecture modifications to improve service level SLAs.Defining SLIs and SLOs around observability for your specific use case.On top of these environmental controls, the workload team should consider: The existing environment should include auditing, monitoring, and alerting by using integrated monitoring and a notification framework. Your environment should consider resilient architectures, cross-region redundancies, service levels indicators (SLIs), service-level objectives (SLOs), service-level agreements (SLAs), and critical support. IoT applications are unique as they're distributed at massive scale, and operate over unreliable networks with no persistent access or visibility into the end-to-end data flows. The key considerations are how quickly you can detect change and how quickly you can resume operations. Because of this, you should design your architecture with availability and resiliency in mind. ![]() Everything has the potential to break and IoT workloads are no exception.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |