7.09.2025 состоялся релизPithon 3.14!
На фоне хайпа про NoGIL всё позабыли про другие фичи. Особенно про Multiple Interpreters, который обещает изоляцию процессов но с эффективностью потоков! На сколько действительно это будет эффективно мы узнаем позже, потому что сейчас это лишь первый релиз с ограничениями и недоработками.
Но что там про NoGIL? Теперь этот режим не экспериментальный, а официально поддерживаемый, но опциональный.
Чтобы запустить без GIL нужна специальная сборка. И перед стартом нужно объявить переменную PYTHON_GIL=0
Для вас я собрал готовый репозиторий где достаточно запустить скрпит, который всё сделает:
▫️ соберет релизный Python 3.14 в новый Docker-образ
▫️ запустит тесты в контейнере (GIL, NoGIL, MultiInterpreter)
▫️ распечатает результаты
Тест очень простой, усложняйте сами)
Вот какие результаты у меня:
=== Running ThreadPoolExecutor GIL ON
TOTAL TIME: 45.48 seconds
=== Running ThreadPoolExecutor GIL OFF
TOTAL TIME: 6.14 seconds
=== Running basic Thread GIL ON
TOTAL TIME: 45.54 seconds
=== Running basic Thread GIL OFF
TOTAL TIME: 4.74 seconds
=== Running with Multi Interpreter
TOTAL TIME: 18.30 seconds
Если сравнивать GIL и NoGIL, то на мои 32 ядра прирост х7-x10 (почему не х32? 🤷). При этом нам обещают что скорости будут расти с новыми релизами.
Режим без GIL похож (визуально) на async, тоже параллельно, тоже не по порядку. Но это не IO! и от того некоторый диссонанс в голове 😵💫, нас учили не так!
Интересно, что чистый Thread работает быстрей чем ThreadPoolExecutor без GIL.
Ну и где-то плачет один адепт мульти-интерпретаторов😭 Теперь нужно искать где они могут пригодиться с такой-то скоростью. Скорее всего своя область применения найдется.
Отдельно я затестил память и вот что вышло на 32 потока:
ThreadPoolExecutor GIL ON
305.228 MB
ThreadPoolExecutor GIL OFF
500.176 MB
basic Thread GIL ON
90.668 MB
basic Thread GIL OFF
472.444 MB
with Multi Interpreter
1267.788 MB
Пока не знаю как к этому относиться)
В целом - радует направление развития!
#release
#rust#consensus_protocol#cryptocurrency#rust#wasm
Kaspa's Rust implementation boosts speed and security, allowing faster transactions (up to 10 blocks per second) while maintaining decentralization. The recent Crescendo upgrade enhances performance, and the code is open for collaboration, offering a reliable, efficient system for users to transact quickly and securely.
https://github.com/kaspanet/rusty-kaspa
#javascript#cheerp#cheerpx#cpp#lwip#repl#tailscale#vm#wasm#webassembly#webvm#xterm_js
WebVM lets you run a full Linux system directly in your web browser without needing a server. It uses a special engine called CheerpX to safely run unmodified Linux programs by converting x86 code to WebAssembly. You get a real Debian Linux environment with many tools, and it supports networking through Tailscale VPN, so your browser VM can connect securely to the internet. You can also customize and deploy your own WebVM easily using GitHub, making it great for development, testing, or learning Linux without installing anything. This means you can have a powerful, private Linux machine anytime, anywhere, just in your browser[1][2][3].
https://github.com/leaningtech/webvm
#kotlin#android#awt#compose#declarative_ui#desktop#gui#ios#javascript#kotlin#multiplatform#reactive#swing#ui#wasm#web#webassembly
Compose Multiplatform is a Kotlin-based framework by JetBrains that lets you build user interfaces for multiple platforms—iOS, Android, desktop (Windows, macOS, Linux), and web—using mostly shared code. It is based on Jetpack Compose for Android, so you can use similar APIs across platforms, speeding up development and ensuring consistent UI design. iOS support is in beta, web is in alpha, and desktop and Android are stable. You can also access native features like camera or maps easily. This helps you save time, reduce bugs, and create apps that work well everywhere with less effort.
https://github.com/JetBrains/compose-multiplatform
#typescript#3d#browser_based#cad#geometry#occ#occt#opencascade#three#threejs#wasm#webassembly#webcad
Chili3D is a free, open-source 3D CAD application that you can use directly in your web browser. It allows you to create and edit 3D models without needing to install any software on your computer. You can make basic shapes like boxes and spheres, draw 2D sketches, and perform advanced operations like combining shapes. It also includes tools for precise alignment and measurement. This makes it easy for anyone to design and edit 3D models online, which is helpful for both beginners and experienced users.
https://github.com/xiangechen/chili3d
#cplusplus#artificial_intelligence#cloud#cloud_native#cncf#container#docker#edge_computing#ewasm#hacktoberfest#hacktoberfest2023#kubernetes#rust_lang#serverless#wasm#webassembly
WasmEdge is a fast, lightweight, and secure WebAssembly runtime that lets you run programs safely on your devices, servers, or the cloud. It supports many programming languages like C++, Rust, and JavaScript, and can run AI models, microservices, and smart contracts efficiently. WasmEdge offers strong security by isolating programs, making it great for extending software safely. It works well on edge devices, smart devices, and cloud environments, and supports easy integration with tools like Kubernetes and Docker. Using WasmEdge helps you run powerful applications faster, safer, and more flexibly on various platforms[1][2][3][4][5].
https://github.com/WasmEdge/WasmEdge
#rust#ai#ai_ocr#attention_mechanism#gnn#gnn_model#gnns#graph#graph_neural_networks#llm_inference#low_latency#mincut#neo4j#ocr#onnx#rust#vector#wasm
RuVector is a free, open-source vector database that gets smarter with every query. Unlike static databases, it learns from usage via GNN layers, runs LLMs locally with no cloud costs, supports graph queries like Neo4j, scales freely across nodes, and deploys as a single self-booting file (125ms startup). Run with `npx ruvector`. You benefit from faster, more accurate AI search that improves automatically, zero operating costs, full offline/privacy control, and easy scaling—perfect for RAG, agents, or edge apps without vendor lock-in.
https://github.com/ruvnet/ruvector
#cplusplus#avx#avx_512#avx_instructions#avx2#avx512#intrinsics#neon#simd#simd_instructions#simd_intrinsics#simd_library#simd_parallelism#simd_programming#sse42#wasm
Highway is a C++ library that helps make software run faster and use less energy. It does this by using SIMD (Single Instruction, Multiple Data) instructions, which let the CPU perform the same operation on many pieces of data at once. This can make programs up to 10 times faster and reduce energy use by up to five times. Highway works on many different types of computers and is easy to use, making it a good choice for developers who want to improve their software's performance.
https://github.com/google/highway