AWS EC2 instances now support ENA queue allocation for your network interfaces

AWS announces a new EC2 feature for Elastic Network Adapter (ENA) that enables flexible queue allocation per Elastic Network Interface (ENI) on EC2 instances. ENA queues, which are key components of ENIs, efficiently manage network traffic by load-balancing sent and received data across available queues. This network interface feature optimizes networking performance by flexibly allocating multiple transmit and receive ENA queues, efficiently distributing packet processing across vCPUs. Customers now have granular control over their network resources and instance performance, allowing them to align ENA queue allocation with specific workload requirements.

Prior to this announcement, customers could configure additional ENIs for their instances, but ENA queues were statically allocated per ENI without flexibility in distribution. Now, customers can dynamically allocate ENA queues across ENIs from their instance's total queue pool, with the total available queues varying by instance type and size. This flexible ENA queue allocation enables maximum vCPU utilization through optimized resource distribution. Network-intensive applications can be allocated more queues, while CPU-intensive applications can operate with fewer queues.

EC2 Flexible Queues is available in all AWS Commercial Regions. To learn more and for supported instance types, review the latest EC2 Documentation.