Skip to main content

> SERVERLESS

Serverless Asynchronous Event & Queue Processing

Decoupled event pipeline using Amazon EventBridge, SQS FIFO queues, and batched Lambda consumers with dead-letter queue resilience.

Back to Architecture CatalogSize This Workload

Mathematical Breakeven Inflection Curve

Cost-optimal up to 50M events/month. Beyond 100M events/month, self-hosted Kafka or managed RabbitMQ on EC2 lowers cost by 65%.

3 Maturity Tiers & Infrastructure Specifications

Component stack and cost steps from prototype to hyper-scale enterprise

1. Prototype / Early Stage0 - 5M events/mo
$10 - $60 / mo
$0.012 / 1K events
Stack Components:
  • EventBridge Default Bus
  • Amazon SQS Standard
  • Lambda Batch Consumer (size=10)
Cost Allocation:Account-level Tagging
Autoscaling:Queue Depth Metric Scaling
2. Scaled Production5M - 50M events/mo
$60 - $550 / mo
$0.010 / 1K events
Stack Components:
  • EventBridge Custom Bus
  • Amazon SQS FIFO
  • Lambda Batch Consumer (size=50) + DLQ
Cost Allocation:Cost Center Tagging per event topic
Autoscaling:SQS ApproximateNumberOfMessagesVisible Autoscaler
3. High-Throughput Enterprise50M - 250M events/mo
$550 - $2,400 / mo
$0.009 / 1K events
Stack Components:
  • EventBridge Pipes
  • Amazon SQS FIFO High-Throughput
  • Lambda Reserved Concurrency Pool
Cost Allocation:Granular FOCUS Tagging per producer service
Autoscaling:EventBridge Pipes Filter-Level Autoscaling

Key Cost Drivers

  • •EventBridge Custom Events ($1.00/M events)
  • •SQS Requests ($0.40/M requests)
  • •Lambda batch compute time

Waste Vulnerabilities

  • •Batch size set to 1 causing excessive Lambda invocations
  • •Unprocessed poisonous messages looping indefinitely

Mitigation Playbooks

  • •Enforce BatchSize=50 or 100
  • •Configure MaximumBatchingWindowInSeconds=5
  • •Set SQS maxReceiveCount=3