KernelScript tackles the growing complexity of eBPF projects by unifying kernel‑side programs, userspace loaders, and kernel modules into a single codebase, using annotations to let the compiler generate the necessary glue code, thereby reducing boilerplate and improving team productivity.
This article walks through removing the per‑CPU map used to pass the TCP‑option offset in an XDP program, shows the required code changes, explains the verifier errors that arise, and presents the final fix using an int offset with a bitwise mask.
Nickel is a high‑performance, XDP‑based four‑layer edge load balancer that replaces Bilibili’s non‑clustered CDN architecture with a co‑deployed control plane and data plane, using hash‑based forwarding, dynamic health‑checks, and two‑hop routing to lower cost, improve load distribution, and eliminate dedicated load‑balancer machines.
This article examines XDP (eXpress Data Path), a Linux kernel eBPF‑based technology that pushes packet processing to the earliest point in the network interface, delivering ultra‑low latency, enhanced security, and flexible custom processing for high‑performance routing, DDoS mitigation, and cloud environments.
XDP (eXpress Data Path) is a Linux kernel eBPF hook that enables ultra‑fast packet processing at the driver level, offering multiple action codes, three deployment modes, and advantages over DPDK such as lower CPU usage, security, and seamless integration with existing networking tools.
This article reviews the OSDI20 best paper on hXDP, explaining why combining XDP with FPGA NICs can alleviate CPU bottlenecks, describing the challenges of eBPF offload to low‑frequency FPGA, and summarizing the custom instruction‑set optimizations and test results that achieve comparable performance to multi‑GHz CPUs.
This article explains how eBPF evolved from classic packet filtering to a powerful, programmable kernel virtual machine that enables high‑performance networking, hardware offload, and flexible security, making it increasingly vital for operators, system administrators, and enterprise solution providers.
This article explains Kubernetes' core networking model, details the various Service types (PodIP, HostPort, NodePort, ExternalIP, LoadBalancer, ClusterIP), describes Cilium's eBPF/XDP implementation for high‑performance load balancing, and presents performance benchmarks and recent BPF kernel extensions.
This article compares four classic Linux packet‑capture engines—libpcap (and its mmap variant), PF_RING, DPDK, and XDP—explaining their data‑flow architectures, copy operations, zero‑copy techniques, and typical use‑cases such as DDoS defense and high‑speed networking.
This article compares four classic Linux packet‑capture engines—libpcap, libpcap‑mmap, PF_RING, DPDK, and XDP—detailing their data‑path architectures, copy‑count reductions, zero‑copy techniques, and typical use‑cases such as DDoS defense and high‑speed trading.
This article explains the fundamentals of cloud native computing, introduces service mesh architectures such as Istio and Envoy, explores chaos engineering with Chaos Mesh, and details how eBPF/XDP‑based user‑space container networking can accelerate data‑plane performance in modern microservice environments.
This article examines the limitations of traditional L4 load balancers like LVS and explains how modern software solutions—leveraging DPDK, VPP, XDP/eBPF, and consistent‑hash algorithms such as Google Maglev and Facebook Katran—can achieve high‑throughput, scalable, and highly available L4 load balancing in data‑center environments.
