11 dépôts
Tools for analyzing bundles, optimizing change detection, and improving runtime speed.
Explore 11 awesome GitHub repositories matching part of an awesome list · Performance and Optimization. Refine with filters or upvote what's useful.
PostCSS is a CSS post-processor and abstract syntax tree transformation tool that parses stylesheets into a structured tree for programmatic analysis and modification. It functions as a plugin-driven pipeline where JavaScript plugins can modify, insert, or delete nodes to transform styles. The project provides a framework for building a custom plugin ecosystem to extend the CSS language with non-standard features such as loops, conditionals, and shorthand properties. It supports multi-syntax parsing through pluggable parsers and stringifiers, allowing it to process various style formats and c
Reduces file size and improves load times by removing unused styles and minifying variables.
Graal is a compiler and runtime architecture designed for high-performance execution and polyglot interoperability. It utilizes a graph-based representation of program logic to perform global optimizations and JIT compilation. The project features a meta-circular interpretation framework and a specialized partial evaluation mechanism, which allow for the creation of new programming languages and the automatic optimization of their semantics into machine code. It enables multiple diverse programming languages to share memory and communicate through a standardized cross-language protocol within
Improves application execution speed and resource efficiency through advanced compiler graph analysis and optimization.
vue-admin-template is a boilerplate for building administrative interfaces using Vue.js. It provides a pre-configured layout, routing, and state management to bootstrap admin dashboards. The project includes a role-based access control system to restrict menu visibility and route access based on user permissions. It features a component scaffolding tool for generating boilerplate views and business components from templates, as well as an automated SVG icon management system for registering and rendering vector graphics. The template covers data management through Excel import and export uti
Implements a performance optimization suite for analyzing bundle sizes and optimizing SVG assets.
This project is a comprehensive functional programming curriculum and learning resource for Haskell. It provides sequenced educational paths and technical reference guides designed to take developers from beginner to advanced levels of proficiency. The project distinguishes itself through a deep focus on theoretical and technical foundations, offering detailed studies on type theory, category theory, and runtime internals. It includes a dedicated performance handbook for optimizing execution speed and memory management, as well as an ecosystem guide for managing development tools and editor c
Provides a handbook for optimizing runtime performance through core representation analysis and static binary linking.
SD.Next is an all-in-one web interface and multi-backend inference engine for generating, editing, and processing images and videos using diffusion models. It functions as a comprehensive tool for diffusion model management and an automated image processing pipeline for bulk operations. The project is distinguished by its hardware-backend abstraction layer, which provides automatic detection and acceleration for NVIDIA CUDA, AMD ROCm, Intel OpenVINO, and DirectML. It features a headless generative API and a programmatic command interface, allowing users to trigger tasks via REST API or CLI wi
Allows tuning of MIOpen environment variables to optimize the trade-off between startup speed and inference performance.
This is a Go backend template that structures a web service into domain, usecase, controller, and repository layers with strict dependency inversion. It provides a foundation for building maintainable and testable REST APIs by separating business logic from transport and data access concerns. The project implements JWT-based authentication, issuing access and refresh tokens for user signup, login, and protected endpoint access. It uses the Gin HTTP framework to build a Docker-packaged REST API with public and private route groups, request validation, and middleware-based authentication. Depen
Analyzes bundles, optimizes change detection, and improves runtime speed.
NCCL est une bibliothèque de communication haute performance et un framework de calcul GPU distribué conçu pour exécuter des échanges de données collectifs et point à point sur plusieurs GPU dans des systèmes à un ou plusieurs nœuds. Il sert de couche de transport GPU RDMA et d'orchestrateur de mémoire, facilitant la synchronisation à large bande passante des données et des gradients de modèle pour l'entraînement et l'inférence GPU distribués. La bibliothèque se distingue par sa capacité à exécuter des primitives de communication directement depuis les noyaux (kernels) GPU, supprimant le CPU hôte du chemin critique. Elle utilise une sélection de chemin consciente de la topologie pour optimiser le mouvement des données et emploie un transport réseau basé sur RDMA, incluant InfiniBand et NVLink, pour permettre un accès mémoire zéro-copie entre les appareils sur différents nœuds physiques. Le projet couvre un large éventail de modèles de communication collective, notamment les réductions, les diffusions (broadcasts), les rassemblements (gathers) et les échanges tous-à-tous, ainsi que l'accès mémoire distant point à point. Il fournit une gestion complète des communicateurs pour initialiser, partitionner et redimensionner les groupes GPU, ainsi qu'une gestion spécialisée de la mémoire pour enregistrer les tampons (buffers) et coordonner la mémoire partagée des appareils. Le système inclut une suite d'outils de surveillance et d'observabilité pour le suivi de la santé, la journalisation diagnostique et la surveillance des événements en temps réel, ainsi que des interfaces d'intégration pour les frameworks de machine learning, les graphes CUDA, MPI et Python.
Configures execution behavior, network module selection, and kernel resource allocation for collective groups.
Ce projet est un framework de traitement de données tabulaires haute performance pour R, conçu pour gérer des jeux de données massifs avec efficacité mémoire et vitesse. Il fournit une structure de données améliorée qui utilise la sémantique de référence et la modification sur place pour effectuer des transformations complexes sans la surcharge de copies d'objets inutiles. La bibliothèque se distingue par ses optimisations architecturales de bas niveau, incluant le traitement parallèle multi-threadé, le tri basé sur radix et l'analyse de fichiers mappés en mémoire. En déchargeant les routines critiques de manipulation et d'agrégation de données vers du code C compilé, elle permet une exécution rapide des tâches qui seraient autrement coûteuses en calcul. Son moteur principal prend en charge des opérations relationnelles avancées, telles que les jointures non-équi, glissantes et à intervalles chevauchants, parallèlement à l'indexation secondaire automatique pour accélérer l'accès répété aux données. Au-delà de ses capacités de traitement principales, le projet offre une suite complète d'outils pour la gestion du cycle de vie des données. Cela inclut des utilitaires d'ingestion et de sérialisation à haute vitesse avec détection automatique de type, ainsi qu'un support spécialisé pour l'analyse de séries temporelles et l'agrégation multidimensionnelle. Le framework est conçu pour évoluer, permettant aux utilisateurs d'effectuer des opérations complexes de regroupement, de filtrage et de remodelage sur des jeux de données contenant des milliards de lignes tout en maintenant la stabilité et les performances du système.
Accelerates grouping, rolling calculations, and transformations using high-performance internal execution paths.
This project is a CUDA programming course and technical guide focused on writing and optimizing GPU kernels for hardware acceleration. It provides structured learning resources for using the CUDA platform to execute operations on silicon architectures. The material covers the optimization of linear algebra kernels and the analysis of machine learning deployment. It includes guidance on identifying acceleration tools, mapping the deep learning ecosystem, and evaluating the frameworks used to move models from research to production environments. The scope extends to GPU performance optimizatio
Guides the improvement of execution speed by fusing linear algebra operations and generating optimized machine code.
cuda-python provides low-level Python bindings for the CUDA Driver and Runtime APIs. It serves as a programmatic wrapper for controlling device memory, managing hardware toolchains, and orchestrating execution graphs on NVIDIA GPUs, allowing for the compilation and launching of parallel kernels directly from Python. The project enables the development of SIMT kernels and the execution of mathematical algorithms on device memory. It integrates pre-compiled bytecode as custom operators and interfaces with accelerated device libraries to access low-level hardware functions without leaving the la
Optimizes GPU execution speed using techniques like cooperative reductions and bytecode caching.
Grule is a business rule engine for Go that decouples complex decision-making logic from core application code. It provides a framework for defining, versioning, and executing business rules through a domain-specific language, allowing logic to be managed independently of the underlying software implementation. The engine distinguishes itself by utilizing a formal grammar-based parser and a Rete-inspired pattern matching algorithm to evaluate logic against data facts efficiently. It supports dynamic system modeling by enabling runtime updates to policies and providing thread-safe knowledge ba
Applies cycle detection and algorithmic shortcuts to ensure fast and efficient rule evaluation.