On July 13, 2006, AWS launched Amazon Simple Queue Service (Amazon SQS) as one of the first three services available to customers, alongside Amazon EC2 and Amazon S3. The service was designed around a requirement in distributed systems: components needed a reliable way to pass messages without creating tight dependencies. Instead of calling another service directly, a producer could place a message in a queue while a consumer processed it when ready. This asynchronous model helped keep failures in one service from affecting the rest of the system.

Twenty years later, decoupling producers from consumers remains the core reason customers use SQS. The service has continued to expand its scale, performance, security, and operational controls. AWS stated that the first 15 years included milestones such as the original 8 KB message limit in 2006, FIFO queues, server-side encryption, and Lambda integration. From 2021 through 2026, further updates addressed throughput, access control, message recovery, integration, and workload size.

Milestones from 2021 through 2025

  • High throughput mode for FIFO queues launched generally in May 2021 at up to 3,000 transactions per second per API action. The limit later increased to 6,000 TPS in October 2022, 9,000 TPS in August 2023, 18,000 TPS in October 2023, and 70,000 TPS per API action in select Regions by November 2023.
  • SSE-SQS launched in November 2021, and became the default for newly created queues in October 2022. The option uses Amazon SQS-managed encryption keys and requires no key management.
  • Dead-letter queue redrive capabilities expanded from the SQS console in December 2021 to the AWS SDK and CLI in June 2023, with FIFO queue support added in November 2023.
  • Attribute-based access control arrived in November 2022, allowing permissions to be configured using queue tags. JSON protocol support, added in November 2023, reduced end-to-end message processing latency by up to 23% for a 5 KB payload.
  • SQS console integration with Amazon EventBridge Pipes was added in 2023. The Extended Client Library became available for Python in 2024, supporting messages up to 2 GB by storing payloads in Amazon S3 and passing a reference through the queue.
  • The FIFO in-flight message limit increased from 20,000 to 120,000 in 2024. In 2025, fair queues addressed noisy-neighbor concerns in multi-tenant standard queues, and the maximum message payload increased from 256 KiB to 1 MiB for standard and FIFO queues.

The underlying use cases remain consistent despite these additions. Customers use SQS to decouple services, buffer bursts of traffic, and build systems that remain resilient when individual components fail. The same pattern also applies to AI workloads, where customers use queues to buffer requests to large language models, manage inference throughput, and coordinate communication between autonomous AI agents operating as independent services.

AWS Lambda event source mapping for SQS was updated in parallel to support the 1 MiB payload size. AWS directs customers seeking more information to the Amazon SQS product page, the developer guide, and recent AWS Blog updates.