FastClick and Beyond: A Look at High-Speed Software Dataplanes and Their Upcoming Challenges
summary
Covers building 100-Gbps-capable network functions (router, load‑balancer, firewall, NAT, DPI, IDS) using FastClick with DPDK or Netmap, and surveys the evolution of graph-based software dataplanes from Click to VPP and BESS. Explains the challenge that multi‑hundred‑Gbps packet rates make memory accesses (L3 cache time) a bottleneck, and presents PacketMill: tighter DPDK integration to shrink memory footprint plus a pipeline and compiler-like graph-derived data structures to cut instruction counts, with these optimizations applied to FastClick.
AI-written summary · about
related talks
29mEnabling Fast, Dynamic Network Processing with ClickOS
16mA High-Speed Stateful Packet Processing Approach for Tbps Programmable Switches
02h 53mPreview Sessions
25mDPDK performance optimization for edge computing scenario
31mFlow-based Packet Processing Framework on DPDK and VPP
36mRunning High-performance User-space Packet Processing Apps in Kubernetes
19mEnabling Programmable Transport Protocols in High Speed NICs
17mPerformance Interfaces for Network Functions