Technology Category
Search and Optimization: Distributed SearchPlanning, Routing, and Scheduling: Optimization of Spatio-temporal SystemsMachine Learning: Optimization
Some optimization possibilities in data plane programming
🤖 AI Summary
To address performance bottlenecks—such as low packet-processing throughput and suboptimal link utilization—in data-plane programming (particularly P4) for SDN, this paper proposes a four-dimensional co-optimization framework: (1) extending the P4 language to support asynchronous external function calls; (2) designing a lightweight, load-size-adaptive compression mechanism; (3) introducing in-network caching to reduce redundant traffic overhead; and (4) offloading high-overhead functions to multi-threaded hosts, VMs, or remote servers. Experimental evaluation demonstrates that the approach preserves full P4 programmability while significantly improving throughput (average +37%), reducing end-to-end latency (up to −42%), and enhancing utilization of both network links and computational resources. This work establishes a systematic, deployable optimization framework for high-performance, highly flexible data-plane programming.
Technology Category
Application Category
📝 Abstract
Software-defined networking (SDN) technology aims to create a highly flexible network by decoupling control plane and the data plane and programming them independently. There has been a lot of research on improving and optimizing the control plane, and data plane programming is a relatively new concept, so study on it is one of the hot topics for researchers. At the 2019 Dagstuhl Seminar, well-known scientists on computer networking discussed challenges and problems in the field of data plane programming that need to be addressed over the next 10 years. Based on this seminar issues and papers review, we suggested some possible solutions which are for optimizing data plane to improve packet processing performance and link utilization. The suggestions include (i) enriching data plane language with asynchronous external function, (ii) compression based on payload size, (iii) in-network caching for fast packet processing, and (iv) offloading external functions to an additional thread, virtual machine (VM) or server, etc. In addition, we implemented some of these in the P4 data plane language to illustrate the practicality.
Problem
Research questions and friction points this paper is trying to address.
Optimizing data plane programming for better packet processing
Enhancing link utilization in software-defined networking
Addressing challenges in data plane language functionality
Innovation
Methods, ideas, or system contributions that make the work stand out.
Enriching data plane with asynchronous functions
Compression based on payload size
Offloading functions to VM or server
🔎 Similar Papers
💼 Related Jobs
