AWSAmazon MQApplication Integration

`aws_mq_broker` cost estimation

A managed ActiveMQ or RabbitMQ message broker. Priced per broker-instance-hour by instance type and deployment mode, plus storage.

An aws_mq_broker runs managed ActiveMQ or RabbitMQ so you don't operate the broker yourself. The cost is per broker-instance-hour, set by the instance type (mq.t3.micro up to mq.m5.4xlarge) and multiplied by the number of instances the deployment mode implies.

Deployment mode is the multiplier people miss. SINGLE_INSTANCE is one broker instance. ACTIVE_STANDBY_MULTI_AZ runs two (a hot standby in another AZ), so it costs twice the instance-hour. RabbitMQ cluster deployments run three. c3x reads host_instance_type and deployment_mode and multiplies the instance-hour rate accordingly.

Storage is the second dimension. ActiveMQ on EBS bills per GB-month for the broker's message store; the throughput-optimized option costs more. c3x prices the instance-hours from the configuration; storage is included where the size is declared, and message volume itself is not separately metered by Amazon MQ.

Terraform example

A minimal but realistic configuration that C3X can estimate.

resource "aws_mq_broker" "events" {
  broker_name        = "events"
  engine_type        = "ActiveMQ"
  engine_version     = "5.18"
  host_instance_type = "mq.m5.large"
  deployment_mode    = "ACTIVE_STANDBY_MULTI_AZ"

  user {
    username = "admin"
    password = var.mq_password
  }
}

Pricing dimensions

What you actually pay for when you provision aws_mq_broker.

Dimension	Unit	What's being charged
Broker instance hours	per instance-hour	Rate for the host_instance_type, multiplied by the instance count from the deployment mode (1 single, 2 active/standby, 3 RabbitMQ cluster). c3x assumes 730 hours/month. $0.30/hour for mq.m5.large ActiveMQ
Storage	per GB-month	Broker message-store storage on EBS. Throughput-optimized storage costs more than the standard tier. $0.10/GB-month (EBS)

Sample C3X output

Example output from c3x estimate (active/standby = 2 instances):

aws_mq_broker.events
└─ Broker instances (mq.m5.large x 2)   1,460  instance-hours    $438.00

OVERALL TOTAL                                                   $438.00

Optimization tips

Common ways to reduce aws_mq_broker cost without changing the workload.

Use SINGLE_INSTANCE for non-production

50%

Active/standby doubles the instance-hour cost for the HA standby. Dev and test brokers rarely need it; SINGLE_INSTANCE halves the bill.

Right-size the instance type

Up to 80% at small sizes

Brokers are often over-provisioned. mq.t3.micro handles low-throughput queues for a fraction of an m5. Move up only when connection or throughput limits bind.

Consider SQS/SNS for simple queueing

Workload-dependent

If you don't need JMS/AMQP protocol compatibility, aws_sqs_queue and aws_sns_topic are pay-per-request with no always-on broker, often far cheaper for moderate volume.

FAQ

How does c3x estimate Amazon MQ cost?

It reads host_instance_type and deployment_mode, prices the broker instance-hour at 730 hours/month, and multiplies by the instance count the mode implies (1, 2, or 3). Storage is added from the declared size.

Why does active/standby cost twice as much?

ACTIVE_STANDBY_MULTI_AZ provisions a second broker instance in another AZ for failover. Both instances bill at the full instance-hour rate, so the deployment costs 2x a single instance.

Are ActiveMQ and RabbitMQ priced the same?

The instance-hour rates are the same per instance type, but RabbitMQ cluster deployments run three instances, so a clustered RabbitMQ broker costs 3x a single instance.

Does message volume affect the cost?

No. Amazon MQ bills for the broker instance and storage, not per message. Throughput is bounded by the instance type you pick, not metered separately.

Is this cheaper than SQS?

Usually not for simple workloads. SQS has no always-on cost and bills per request. Amazon MQ is worth its standing cost only when you need JMS, AMQP, MQTT, or STOMP protocol compatibility.