The trigger (powered by Event Hubs) ensures that only one EventProcessorHost instance can get a lease on a given partition.įor example, consider an event hub with the following characteristics: This section covers how Azure Functions responds to events sent to an event hub event stream using triggers.Įach instance of an Event Hubs triggered function is backed by a single EventProcessorHost instance. Consuming events with Azure FunctionsĪzure Functions supports trigger and output bindings for Event Hubs. To learn more, see Deep dive on Event Hubs concepts and features. Consumers and consumer groupsĪ partition can be consumed by more than one consumer, each reading from and managing their own offsets.Įvent Hubs has the concept of consumers groups, which enables multiple consuming applications to each have a separate view of the event stream and read the stream independently at their own pace and with their own offsets. This behavior provides multiple degrees of parallelism and enhances throughput for consumers. When more than one partition is used, it allows for parallel logs to be used from within the same event hub. The analogy of a commit log is frequently used to describe the nature of how events are added to the end of a sequence in a partition. Each partition works as a log where records are written in an append-only pattern. Each event is written in exactly one partition and isn't multi-cast across partitions. When a partition isn't specified by the sender, received events are distributed across partitions in the event hub. Events are immutable and persisted in an event hub, also referred to as a topic in Kafka. The following diagram shows the Event Hubs stream processing architecture:Īn event is a notification or state change that is represented as a fact that happened in the past. Before diving into the patterns and best practices for Azure Functions integration, it's best to understanding the fundamental components of Event Hubs. Event Hubs core conceptsĪzure Event Hubs is a highly scalable event processing service that can receive millions of events per second. This article provides guidance on how to effectively take advantage of this integration by highlighting key considerations and techniques for performance, resiliency, security, observability, and scale. As much as these services work seamlessly together, there are many features, settings, and intricacies that add complexity to their relationship. Solutions that use Azure Event Hubs together with Azure Functions benefit from a serverless architecture that is scalable, cost-effective, and capable of processing large volumes of data in near real time.

Event hub