Topic

#Performance

31 articles on Performance — news, releases, guides and analysis from the SourceFeed engine.

Article 4d ago 4

Giant JSON Files Hit a Wall Long Before RAM Runs Out

Node's hard-coded string cap, not your heap, is the ceiling — and the real fix sits upstream of any parser.

Priya Nair

Why FreeBSD Appears to Eat Your RAM (and Why It Doesn't Matter)

Demystifying FreeBSD's page queues, ZFS ARC, and why your monitoring tools are lying to you.

Article · 1mo ago1

Stop the Linux OOM Killer From Wrecking Your PostgreSQL Database

Taming the Linux virtual memory manager with strict overcommit is the only way to prevent silent database crashes.

Article · 1mo ago2

Fixing the DHT: How IPFS Achieved Sub-Second Publishing

By replacing rigid deterministic lookups with statistical heuristics, Kubo 0.39.0 slashes content routing latency from twenty seconds to one.

Article · 1mo ago0

Ditching std::unordered_map for Cache-Friendly Hopscotch Hashing

Why Tessil's hopscotch-map library beats standard C++ hash tables in memory efficiency and raw lookup speed.

Article · 2mos ago7

Architecting the Next.js App Router Migration for Speed

How transitioning from imperative client-side fetching to a hybrid server-client model slashes latency and bundle sizes.

Article · 2mos ago3

Profile and Fix a Slow Python Service Using py-spy Flame Graphs

Attach py-spy to a live process with zero code changes, read the flame graph to pinpoint an O(n²) bottleneck, and verify a 1000x+ speedup — all without restarting your service.

Tutorial · 2mos ago0

Beyond Epoll: Is io_uring Ready for Production?

A technical breakdown of Linux event loops, comparing readiness-based epoll with completion-based io_uring for high-performance networking.

Article · 2mos ago0

The Context Trap: Why Headroom’s Local Compression Layer is Essential for AI Agents

Infinite context windows are a costly illusion; local, content-aware compression is the missing optimization layer for production AI.

Article · 2mos ago2

Java’s Identity Crisis Ends: Project Valhalla Lands in JDK 28

Value classes finally arrive as a preview in JDK 28, bringing bare-metal performance to Java's object-oriented abstractions.

Article · 2mos ago4

Demystifying Integer Quantization for Neural Network Inference

A low-level look at the mathematics and hardware mechanics that shrink massive models without destroying accuracy.

Article · 2mos ago0

TypeScript 7.0 RC Debuts Go-Based Compiler with 10x Speedups

The Microsoft team ports the compiler to Go, introducing shared-memory parallelism while maintaining structural parity with version 6.0.

Release · 2mos ago0

Writing a C++20 Path Tracer From Scratch Without AI

A dependency-free renderer showcases the power of modern C++20, custom acceleration structures, and classic graphics engineering.

Article · 2mos ago4

Migrating TrueType Hinting to Swift: How Apple Beat C

Apple's rewrite of its legacy TrueType interpreter shows how modern Swift features like non-copyable types can outperform C.

Article · 2mos ago4

The Unreasonable Power of Plain HTML

Why building with basic semantic markup remains the ultimate strategy for resilient, accessible, and universally compatible web applications.

Article · 2mos ago0

How Linear Gets to Sub-10ms UI Updates: The Architecture Behind the Speed

Linear's legendary snappiness isn't magic — it's a deliberate set of architectural decisions. Here's how they work, and which patterns you can steal today.

Article · 2mos ago0