--- title: A Quick Look at AWS Scalable Reliable Datagram Protocol... description: AWS introduced ENA Express at re:Invent, using the Scalable Reliable Datagram (SRD) protocol as the transport layer for overlay virtual networks. SRD is a... canonical: https://daily.dev/posts/a-quick-look-at-aws-scalable-reliable-datagram-protocol-ipspace-net-blog-utddzpwf1 twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: A Quick Look at AWS Scalable Reliable Datagram Protocol « ipSpace.net blog | daily.dev og:description: AWS introduced ENA Express at re:Invent, using the Scalable Reliable Datagram (SRD) protocol as the transport layer for overlay virtual networks. SRD is a... og:url: https://daily.dev/posts/a-quick-look-at-aws-scalable-reliable-datagram-protocol-ipspace-net-blog-utddzpwf1 og:image: https://api.daily.dev/og/posts/UtdDZpWf1.png og:image:alt: A Quick Look at AWS Scalable Reliable Datagram Protocol « ipSpace.net blog og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # A Quick Look at AWS Scalable Reliable Datagram Protocol « ipSpace.net blog **[ipSpace.net Blog](https://daily.dev/sources/ipspace)** · 4 min read · 0 upvotes · 0 comments ## Summary AWS introduced ENA Express at re:Invent, using the Scalable Reliable Datagram (SRD) protocol as the transport layer for overlay virtual networks. SRD is a reliable datagram protocol that, unlike TCP, allows out-of-order packet delivery, enabling packets to be spread across multiple parallel paths simultaneously. This multipath spreading boosts single TCP flow throughput to 25 Gbps and reduces tail latency (99.9th percentile) by 85% for high-throughput workloads. The key latency improvement comes from eliminating TCP timeout penalties caused by dropped tail packets — a problem where the sender must wait milliseconds for a timeout while fabric transit times are in microseconds. The post debunks hype around DPUs being required for this, noting SRD could be implemented in a virtual switch, and contextualizes SRD as a well-understood concept with historical precedents like X.25 and V.42. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### Why does AWS ENA Express with SRD reduce tail latency for TCP workloads? The primary cause of TCP tail latency is losing the last packet of a burst — the receiver cannot send a Selective ACK because no subsequent packet arrives, forcing the sender to wait for a timeout. On most operating systems the minimum TCP timeout is a few milliseconds, while fabric transit times are measured in microseconds. SRD provides reliable underlay transport, eliminating these timeout-driven stalls and cutting 99.9th-percentile latency by 85%. _Engineers tuning high-throughput EC2 workloads track ENA Express developments like this on daily.dev._ ### How does AWS SRD protocol differ from TCP and UDP? SRD is a reliable datagram protocol that combines properties of both TCP and UDP. Like UDP it provides datagram transport; like TCP it guarantees reliable delivery. Unlike TCP, SRD allows packets to be delivered out of order, enabling them to be sent in parallel across all available network paths simultaneously. Consumers of SRD must handle packet reordering themselves, though ENA Express reorders incoming SRD packets before passing them to the VM TCP/IP stack. _Developers building on AWS networking primitives find protocol-level breakdowns like this on daily.dev._ ## Similar posts on daily.dev - [Tolerating Delay With DTN](https://daily.dev/posts/tolerating-delay-with-dtn-xuicwfwnh) · Hackaday · 0 upvotes · 0 comments - [Improving NAT traversal, pt. 3: looking ahead](https://daily.dev/posts/improving-nat-traversal-pt-3-looking-ahead-sd7jm9lvd) · Tailscale · 0 upvotes · 0 comments - [MRC and SRv6: How Foundational Networking Innovations Are Enabling the Next Generation of AI Supercomputers](https://daily.dev/posts/mrc-and-srv6-how-foundational-networking-innovations-are-enabling-the-next-generation-of-ai-superco-zifro0frx) · Cisco · 0 upvotes · 0 comments - [AWS Replaces Fat-Tree Data Center Networks with Random Graph Theory, Cutting Routers by 69%](https://daily.dev/posts/aws-replaces-fat-tree-data-center-networks-with-random-graph-theory-cutting-routers-by-69--rzssjbvtk) · InfoQ · 1 upvotes · 0 comments --- Tags: [#aws](https://daily.dev/tags/aws) [View this post on daily.dev](https://daily.dev/posts/a-quick-look-at-aws-scalable-reliable-datagram-protocol-ipspace-net-blog-utddzpwf1) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"A Quick Look at AWS Scalable Reliable Datagram Protocol « ipSpace.net blog","url":"https://daily.dev/posts/a-quick-look-at-aws-scalable-reliable-datagram-protocol-ipspace-net-blog-utddzpwf1","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/a-quick-look-at-aws-scalable-reliable-datagram-protocol-ipspace-net-blog-utddzpwf1"},"datePublished":"2026-08-11T08:56:10.075Z","dateModified":"2026-08-11T09:30:12.481Z","description":"AWS introduced ENA Express at re:Invent, using the Scalable Reliable Datagram (SRD) protocol as the transport layer for overlay virtual networks. SRD is a...","image":"https://media.daily.dev/image/upload/s--2-1xRawN--/f_auto/v1722860399/public/Placeholder%2011","thumbnailUrl":"https://media.daily.dev/image/upload/s--2-1xRawN--/f_auto/v1722860399/public/Placeholder%2011","isAccessibleForFree":true,"articleSection":"ipSpace.net Blog","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"ipSpace.net Blog","logo":"https://media.daily.dev/image/upload/s--xS7E82MT--/f_auto,q_auto/v1786438303/logos/ipspace?_a=BAMAMicg0","url":"https://daily.dev/sources/ipspace"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/a-quick-look-at-aws-scalable-reliable-datagram-protocol-ipspace-net-blog-utddzpwf1","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"aws","timeRequired":"PT4M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"ipSpace.net Blog","item":"https://daily.dev/sources/ipspace"},{"@type":"ListItem","position":3,"name":"A Quick Look at AWS Scalable Reliable Datagram Protocol « ipSpace.net blog"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/a-quick-look-at-aws-scalable-reliable-datagram-protocol-ipspace-net-blog-utddzpwf1#faq","mainEntity":[{"@type":"Question","name":"Why does AWS ENA Express with SRD reduce tail latency for TCP workloads?","acceptedAnswer":{"@type":"Answer","text":"The primary cause of TCP tail latency is losing the last packet of a burst — the receiver cannot send a Selective ACK because no subsequent packet arrives, forcing the sender to wait for a timeout. On most operating systems the minimum TCP timeout is a few milliseconds, while fabric transit times are measured in microseconds. SRD provides reliable underlay transport, eliminating these timeout-driven stalls and cutting 99.9th-percentile latency by 85%. Engineers tuning high-throughput EC2 workloads track ENA Express developments like this on daily.dev."}},{"@type":"Question","name":"How does AWS SRD protocol differ from TCP and UDP?","acceptedAnswer":{"@type":"Answer","text":"SRD is a reliable datagram protocol that combines properties of both TCP and UDP. Like UDP it provides datagram transport; like TCP it guarantees reliable delivery. Unlike TCP, SRD allows packets to be delivered out of order, enabling them to be sent in parallel across all available network paths simultaneously. Consumers of SRD must handle packet reordering themselves, though ENA Express reorders incoming SRD packets before passing them to the VM TCP/IP stack. Developers building on AWS networking primitives find protocol-level breakdowns like this on daily.dev."}}]} ```