Linux Kernel Journey
The article explains how the traditional mmap_lock read lock creates severe contention and priority inversion for frequent madvise MADV_DONTNEED/FREE calls, and how a per‑VMA locking redesign eliminates this bottleneck, improves performance, and is slated for the next Linux kernel release.
This guide walks you through a complete workflow for profiling CUDA applications, covering GPU performance fundamentals, key metrics, NVIDIA Nsight tools, CUPTI programming, example code, common bottlenecks, and best‑practice recommendations to identify and eliminate performance limits.
The article chronicles Xavier Xia’s iterative patches to the Linux kernel’s contpte_ptep_get() function, showing how early‑exit logic and subsequent refinements ultimately yielded consistent performance gains across diverse dirty/young page table scenarios, backed by benchmark data that convinced skeptical reviewers.
This article explains why cache is critical for computer performance and provides a thorough analysis of the three classic cache mapping strategies—direct‑mapped, fully associative, and set‑associative—detailing their mechanisms, advantages, disadvantages, concrete examples, and guidance on selecting the appropriate method for different system scenarios.
The paper presents Tencent’s TGW cloud gateway architecture, highlighting a 2.9× forwarding performance boost, lossless state migration within 4 seconds, sub‑minute fault detection, multi‑level fault‑tolerance mechanisms, and operational best practices that enable 100 % availability for massive online services.
This article thoroughly examines how JuiceFS, a FUSE‑based distributed file system, aligns with Linux's discretionary and mandatory access control models, implements Unix permissions, POSIX ACLs, special permission bits, capability handling, and SELinux integration, while detailing practical configurations such as SDK access, root‑squash, all‑squash, and optimized ACL storage.
This tutorial explains how to build an eBPF‑based tracing tool that intercepts CUDA runtime API calls via uprobes, captures detailed event data such as memory sizes, transfer directions, kernel launches and errors, and presents it in a readable format for debugging and performance analysis.
The article explains how the Model Context Protocol (MCP) introduces Streamable HTTP as a modern, bidirectional transport that overcomes the one‑way, session‑less limitations of Server‑Sent Events, providing richer data formats, built‑in session management, and better resilience for AI assistants.
This article explains how to measure, analyze, and improve CPU usage in software by examining key metrics, choosing efficient algorithms and data structures, leveraging compiler options, exploiting cache behavior and SIMD vectorization, and applying practical case studies with Linux profiling tools.
This 148‑page PowerPoint offers a thorough, still‑relevant overview of virtualization, detailing its core concepts, hardware mechanisms, software implementations, XEN and KVM architectures, peripheral virtualization, and the relationship to containers and Docker.
