From f79f573cb6cb161196dc559e36e79e1145689313 Mon Sep 17 00:00:00 2001 From: KrisNey-MSFT Date: Thu, 10 Oct 2024 14:47:24 -0700 Subject: [PATCH] Update README.md w/Behavioral Model text --- README.md | 2 ++ 1 file changed, 2 insertions(+) diff --git a/README.md b/README.md index 3c696d7a0..38a8f0961 100644 --- a/README.md +++ b/README.md @@ -9,6 +9,8 @@ We are developing set of APIs and object models describing network services for the cloud, and will work with all cloud providers and enterprise hybrid clouds to develop further functionality. We believe the DASH program describes a comprehensive set of services that are required by the vast majority of clouds. The goal of DASH is to be specific enough for SMART Programmable Technologies to optimize network performance and leverage commodity **hardware** technology to achieve 10x or even 100x stateful connection performance. The technology has multiple applications such as 1) NIC on a host, 2) a Smart Switch, 3) Network Disaggregation, and 4) high performance Network Appliances. Many technology companies have committed to further developing this new open technology and its community. The best minds and practitioners are actively collaborating to optimize performance of the cloud by extending SONiC to include stateful workloads. + +With the development of a Behavioral Model, we hope to push the boundaries of what P4, PNA, BMv2, and P4 DPDK can do, and contribute to shape the future of networking technology. Together these components provide a cohesive environment for developing, validating, and deploying programmable network devices that can operate consistently across different HW and SW platforms. Behavioral models help in understanding, designing, and analyzing the dynamic behavior of a system, ensuring that it meets its functional requirements. We plan to develop scenarios and test cases that verify the system behaves as expected under various conditions. The Behavioral Model can also be used to simulate a system's behavior before actual implementation, helping to identify potential issues early. Future innovations for in-service software upgrades and ultra-high availability for stateful connections will also be developed with the utmost importance.