LAUSR.org creates dashboard-style pages of related content for over 1.5 million academic articles. Sign Up to like articles & get recommendations!

Adaptive SmartNIC Offloading for Unleashing the Performance of Protocol-Oblivious Forwarding

The growth of Internet of Things (IoT) has led to the convergence of heterogeneous networking systems powered by various protocols, and consequently the emergence of protocol-independent packet processing based on… Click to show full abstract

The growth of Internet of Things (IoT) has led to the convergence of heterogeneous networking systems powered by various protocols, and consequently the emergence of protocol-independent packet processing based on programmable data plane (PDP) for IoT. In this work, we study how to leverage the hardware acceleration enabled by offloading flow tables to SmartNIC to improve the performance of software PDP switches based on protocol-oblivious forwarding (POF). We design our SmartNIC offloading system (namely, OVS-POF-TC) based on the Linux kernel traffic classification (TC) system and open vSwitch (OVS), extend OVS to enable the installation of POF-based flow tables (POF-FTs) in a SmartNIC, and design a selective offloading mechanism for purposely offloading heavy-load POF-FTs. Our experimental results indicate that OVS-POF-TC offloads and updates POF-FTs timely, supports runtime programmability, and has improved packet processing throughput by $1.52\times $ and $3.82\times $ , when applying POF-FTs with SetField and AddField to packets, respectively. Moreover, to ensure that OVS-POF-TC can offload and replace POF-FTs adaptively, we formulate two mixed-integer linear programming (MILP) models to, respectively, solve the problems of flow placement and flow replacement, and also design time-efficient heuristics for them.

Keywords: pof fts; ovs pof; oblivious forwarding; protocol oblivious; smartnic offloading; performance

Journal Title: IEEE Internet of Things Journal
Year Published: 2023

Link to full text (if available)


Share on Social Media:                               Sign Up to like & get
recommendations!

Related content

More Information              News              Social Media              Video              Recommended



                Click one of the above tabs to view related content.