Skills
.Work. You
Rest

hpc-patterns

Community

Boost HPC performance with cache patterns.

Software Engineering#cache#SIMD#hpc#threading#C++20#NUMA#memory-allocation
Authorysyecust
Version1.0.0
Installs0

System Documentation

What problem does it solve?

High-performance computing applications often suffer from poor data locality and underutilized hardware. This Skill provides proven patterns to improve cache efficiency, vectorization, threading, and NUMA awareness.

Core Features & Use Cases

  • Cache-Friendly Data Structures: SoA vs AoS, layout-driven performance improvements for particle systems, matrices, and graphs.
  • SIMD Vectorization: Guidance on compiler hints, intrinsics, and portable SIMD strategies for uniform workloads.
  • Memory Management: Aligned allocators and memory pools to reduce fragmentation and improve bandwidth.
  • Thread Parallelism: Thread pools, parallel for patterns, and work-stealing strategies for scalable performance.
  • NUMA-Aware Allocation: Techniques to maximize memory locality on multi-socket systems.
  • Performance Measurement: Simple profiling and benchmarking guidance to quantify improvements.

Quick Start

Apply these patterns to a C++20 HPC project by converting a kernel to a Structure-of-Arrays layout, enabling SIMD pragmas, and integrating a 64-byte aligned allocator and a NUMA-aware allocation strategy.

Dependency Matrix

Required Modules

None required

Components

Standard package

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: hpc-patterns
Download link: https://github.com/ysyecust/everything-claude-code/archive/main.zip#hpc-patterns

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 223,000+ vetted skills library on demand.