Posted On: Nov 26, 2023
ENA Express now supports 58 new instance types ranging from 16 to 192 vCPUs. With this launch, ENA Express is adding support for 20 new memory optimized instances, 15 new compute optimized instances, 15 new general purpose instances, and 8 new storage instances. Before today, ENA Express supported only the largest instance size, totaling 27 instances as small as 64 vCPUs.
ENA Express is a networking feature that uses the AWS Scalable Reliable Datagram (SRD) protocol to improve network performance in two key ways: higher single flow bandwidth and lower tail latency for network traffic between EC2 instances. SRD is a proprietary protocol that delivers these improvements through advanced congestion control, multi-pathing, and packet reordering directly from the Nitro card. ENA Express works transparently to your applications with TCP and UDP protocols.
ENA Express with compute optimized instances is great for file systems and media encoding workloads that need higher single flow bandwidth. ENA Express coupled with general purpose instances works well for large object transactions to and from the database. ENA Express and memory optimized instances deliver improved tail latency for high transactional per second databases. Finally, ENA Express and storage based instances take advantage of the improve tail latencies of ENA Express for fastest reads and writes from memory while local storage is used for high I/O transactions.
ENA Express is available in all AWS Commercial Region and comes at no additional cost. For a complete list of supported instances, please review the latest EC2 Documentation.