Azure Event Hubs is a fully-managed, real-time data ingestion service that is simple, secure, and scalable. Event Hubs lets you stream millions of events per second from any source so you can build dynamic data pipelines and respond to business challenges immediately. Keep data ingestion secure with geo-disaster recovery and geo-replication options.
With Azure Event Hubs for Apache Kafka, you can enable existing Kafka clients and applications to talk to Event Hubs without any code changes, giving you a managed Kafka experience without having to manage your own clusters.
|Basic Edition||Standard Edition3||Professional Edition 1|
|Throughput units (1 MB/s ingress, 2MB/s egress)||￥0.1/hour (about ￥74.40/month)||￥0.19/hour (about ￥141.36/month)||Included|
|Ingress events||￥0.18/million events||￥0.18/million events||Included|
|Capture||￥0.636/hour (about ￥473.184/month)||Included|
|Extended retention4||￥1.02/GB/month (10 TB included per CU)|
What are ingress events and how are they billed?
An ingress event is a unit of data 64 KB or less. Each one is a billable event. Larger messages are billed in multiples of 64 KB. For example, 8 KB is billed as one event, but a 96 KB message is billed as two events.
Events consumed from an event hub, as well as management operations and "control calls" such as checkpoints, are not counted as billable ingress events, but accrue to the throughput unit allowance.
What are throughput units and how are they billed?
The throughput unit is selected specifically by customers through Azure Portal or the Event Hubs’ Management API. The throughput unit will be applied to all Event Hubs in a certain Service Bus namespace, and every throughput unit will assign the namespace the following features:
- 1. Up to 1 MB per second of ingress events (= events send into an event hub), but no more than 1,000 ingress events, management operations, or control API calls per second.
- 2. Up to 2 MB per second of egress events (= events consumed from an event hub).
- 3. Up to 84 GB of event storage (sufficient for the default 24-hour retention period).
How is the throughput unit of Event Hubs implemented?
If the total ingress throughput or the total ingress event rate across all event hubs in a namespace exceeds the aggregate throughput unit allowances, senders will get throttled and receive errors indicating that the ingress quota has been exceeded.
If the total egress throughput or the total event egress rate across all event hubs in a namespace exceeds the aggregate throughput unit allowances, receivers will get throttled and receive errors indicating that the egress quota has been exceeded. Ingress and egress quotas are enforced separately so that no sender can cause event consumption to slow down, nor a receiver can prevent events from being sent into Event Hub.
Note that the throughput unit selection is independent of the number of Event Hub partitions (sometimes referred to as shards in similar systems). While each partition offers a maximum throughput of 1 MB/s, 1,000 event/s ingress, and 2 MB/s egress, there is no fixed charge for the partitions themselves. The charge is for the aggregate throughput units on all event hubs in a namespace. With this, customers can create enough partitions to support the anticipated maximum load for their systems, without incurring any throughput unit charges until the event load on the system actually requires higher throughput numbers, and without having to change the structure and architecture of their systems as the load on the system increases.
Example—Suppose you chose eight throughput units on a namespace, and create a single event hub with 32 partitions. If all partitions in this event hub see even load, each partition gets approximately 0.25 MB/s ingress throughput for a total aggregate throughput of 8 MB/s. If a single partition sees a usage spike to 1 MB/s, while eight other partitions only see half their peak load (0.125 MB/s), no throttling will occur. However, if the load peak of a certain partition exceeds 1 MB/second, the partition will be limited (since every partition has a quota), although the aggregated throughput of all partitions is still less than 8 MB/second.
Is there a limit on the number of throughput units that can be selected?
Yes. Basic and standard tier namespaces can have a maximum of 20 throughput units. You can request a larger number of throughput units for a standard tier by filing a support ticket. Note that the immediate availability of additional capacity above 20 throughput units per Azure subscription is not assured. We recommend that customers anticipating a need for more than 20 throughput units of Event Hub capacity contact us to discuss requirements as early as possible in the planning cycle. Additional throughput units are available in blocks of twenty on a committed purchase basis.
What is the maximum retention period?
We provide a maximum of a seven-day retention period upon general availability. Note that event hubs are not intended as a permanent data store. Retention periods greater than 24 hours are intended for scenarios where it's convenient to replay an event stream into the same systems, for example to train or verify a new machine learning model on existing data.
Is there a charge for retaining Event Hub events for more than 24 hours?
In many cases, yes. If the size of the total amount of stored events exceeds the storage allowance for the number of selected throughput units (84 GB per throughput unit), the size that exceeds the allowance is charged at the regular Azure Blob Storage rate. The storage allowance in each throughput unit covers all storage costs for retention periods of 24 hours (the default), even if the throughput unit is used to the maximum ingress allowance.
How is Event Hub storage size calculated and charged?
The total size of all stored events, including any internal overhead for event headers or on disk storage structures in all event hubs in a namespace is measured throughout the day. At the end of the day, the peak storage size is calculated. The daily storage allowance is calculated based on the minimum number of throughput units that were selected during the day (each throughput unit provides an allowance of 84 GB). If the total size exceeds the daily storage allowance, the excess storage is billed using Azure blob storage rates (at the locally redundant storage rate).
Do brokered connections charges apply to Event Hubs?
There are no connection charges for sending events using HTTP, regardless of the number of sending systems or devices. AMQP connections are metered, but the first 100 concurrent connections are free for every basic Event Hubs namespace, and the first 1,000 concurrent connections per subscription are free for standard Event Hubs. These allowances cover most receive scenarios and many service-to-service scenarios. Brokered connections charges usually only become significant if you plan to use AMQP on a large number of clients, for example to achieve more efficient event streaming or activate two-way communication (IoT command and control schemes). Please refer to the Service Bus Connections Pricing information for details on what constitutes a brokered connection and how they are metered.
How is Event Hubs capture billed?
Event Hubs capture is enabled when any event hub in the namespace has the capture feature enabled. Capture is billed hourly per purchased throughput unit. As the throughput unit count is increased or decreased, Event Hubs capture billing will reflect these changes in whole hour increments.
How does Event Hubs capture affect my egress from Event Hubs?
Event Hubs capture does not impact egress rates for Event Hubs throughput units. You can still read at the full throughput unit rate of 2,000 events per second/2 MBps per throughput unit.
How do storage charges apply for the storage account I select for Event Hubs capture?
Event Hubs capture uses a storage account you provide on a schedule you provide. Because this is your storage account, any usage charges for this storage account will be billed to your Azure subscription. The shorter your capture window the more frequent storage transactions will occur.
Support & SLA
If you have any questions or need help, please visit Azure Support and select self-help service or any other method to contact us for support.
For the “Basic” and “Standard” levels of Event Hubs, we guarantee that, at least 99.9% of the time, applications configured correctly can send or receive messages or perform other operations in Event Hubs. If you want to learn about the details of our server level agreement, please access the Service Level Agreements page.