Astra Streaming annual price plans

Edit online

Choose from either a capacity-based or consumption-based price plan.

The annual price plans for Astra Streaming provide the following benefits:
  • Fully managed, Apache Pulsar compatible streaming.
  • Consumption-based or capacity-based pricing.
  • Build and deploy on AWS, Azure, or Google Cloud.
  • Limited cloud regions for consumption-based pricing.
  • Deploy in all cloud regions with capacity-based pricing.
  • Geo-replication, secure private networking, stream processing with Pulsar functions.
  • Native wire-level protocol support for Apache Kafka, JMS, RabbitMQ.
  • Volume discounts with annual commits.
  • 24x7 enterprise support.
  • Option to add premium support for 15-minute response times and other premium features.

Two pricing options are available:

  • Capacity-based pricing: The price is based on the maximum streaming capacity, payload size, and storage that your Astra Streaming deployment can use. For more information about the metrics used, see Table 1. Capacity-based pricing means that you pay a fixed price for the number of Capacity Units (CUs) that you require to meet your load and performance requirements.

  • Consumption-based pricing: The price is calculated based on your actual usage, such as the size of each streaming write operation. For more information about the metrics used, see Table 2. Consumption-based pricing means that you pay a variable cost, dependent on your usage, with no capacity limit.

To see the available Astra Streaming price plans for your cloud provider and region, follow the link for your cloud provider:

Table 1. Capacity-based pricing metric definitions
Metric Billing Unit Metering Details
Streaming Capacity Unit A Streaming Capacity Unit (SCU) is the deployment unit of streaming capacity capable of supporting user workloads and functions. The minimum deployment size for a deployment is 3 SCUs and additional SCUs can be added as needed if more capacity is required. Calculated as the number of hours that the SCU is available to the user during the billing period.
Egress Egress is based on the payload size of each streaming read. Calculated as the total number of bytes read from topics on an Astra Streaming SCU during the billing period.
Storage block A storage block is an allocation of storage that is used for message processing and archival. A storage block provides up to 128 GiB of storage and additional storage blocks can be added as required. Calculated as the number of 128 GiB blocks of storage that are available to the user during the billing period.
Table 2. Consumption-based pricing metric definitions
Metric Billing Unit Metering Details
Write bandwidth

The write bandwidth is based on the payload size of each streaming write.

RabbitMQ clients will incur a “RabbitMQ write bandwidth” charge in addition to the standard write bandwidth charge.

The minimum message byte size is 512 bytes. Messages smaller than 512 bytes are charged as 512-byte messages.
Read bandwidth

The read bandwidth is based on the payload size of each streaming read.

RabbitMQ clients incur a “RabbitMQ read bandwidth” charge in addition to the standard read bandwidth charge.

The minimum message byte size is 512 bytes. Messages smaller than 512 bytes are charged as 512-byte messages.
Streaming storage

Streaming storage means all the messages that retained in persistent storage.

Storage is calculated based on logical storage consumption. Storage is measured in GiB per month, using the maximum data stored during the month for the usage calculation.

Topic/Partitions

Topic/Partitions means the channel that is used to pass the messages published by producers to the consumers who process those messages.

Topic-partitions are measured per hour.

Function instances

A function instance means a unit used for scaling Pulsar functions or Pulsar IO connectors running on Astra Streaming.

Each function instance is assigned a specific amount of CPU and memory, which can be allocated to a Pulsar function. Each Pulsar function that runs on Astra Streaming has a minimum of one function instance allocated to it.

Function instances are measured per hour.

Function read bandwidth

Function read bandwidth is based on the payload size of each function read.

Function read bandwidth is the egress traffic that is measured in GiB coming out of a function (or connector).

Function write bandwidth

Function write bandwidth is based on the payload size of each function write.

Function write bandwidth is the ingress traffic that is measured in GiB coming into a function (or connector).

Metrics read

The metrics read is based on the payload size of each streaming metrics read.

Metrics reads are measured in GiB.