High speed load balancing from the Linux Kernel
summary
ZEVENET builds a Linux-kernel layer‑4 core using Netfilter via nftables, conntrack helpers and a custom nft module set (random hashing, DNAT/SNAT models) and exposes a user‑space “diamond” that runs health checks over JSON/REST while loading nft rule batches atomically. Tests claim 5–10× faster than LVS (empty conntrack and DSR IPv4 cases) and target multi‑gigabit throughput up to 10 Gbps. Packet flows use nftables ingress hook for a one‑rule UDP fast path and kernel offload for scalability.
AI-written summary · about
chapters (4)
related talks
20mNetKernel: Making Network Stack Part of the Virtualized Infrastructure
34mCan the Linux networking stack be used with very high speed applications?
14mHigh-throughput and Flexible Host Networking for Accelerated Computing
31mA Practical, Research‑Backed Introduction To Cloud‑Native Kernel‑Bypass Networking
42mAccelerating Envoy with the Linux Kernel
19mConntrack offload: why and how
14mZero-overhead Container Networking with eBPF and Netkit
17mRearchitecting the TCP Stack for I/O-Offloaded Content Delivery