scheduler

一篇架构对比文章，分析 Pi Durable 与 OpenCode 的设计差异：Pi Durable 将模型生成、工具调用等序列化为可调度的持久任务，使工作不再绑定于单个进程，而传统会话式 harness（如 OpenCode、Claude Code、Codex）仍依赖进程内的活动 agent 循环。

A technical blog post examining the common complaints surrounding async Rust — including function coloring, async drop, Tokio defaultism, and the scoped task trilemma — while exploring where scheduling responsibility should live in the ecosystem.

Cronhq is a scheduler that ensures exactly-once execution for cron jobs, featuring retries, HMAC-signed webhooks, and heartbeats, built on Rust and Postgres, and is open-source under MIT license.

The article presents an RFC patch set by Jens Axboe for io_uring that proposes handling potentially blocking operations without context switching to improve performance by reducing overhead.

The article investigates improvements to Cargo's scheduler by analyzing its behavior through benchmarks of Rust build tasks and exploring alternative scheduling algorithms.

Redox OS has implemented the EEVDF scheduler, replacing DWRR, resulting in significant improvements in fairness, context switch time, and throughput.

Linux 7.2 has been released with significant improvements including cache-aware scheduling, power management for Raspberry Pi GPUs, and various performance enhancements.

Explains why Tokio's scheduler does not preserve task ordering at scale, using a real-world example of 1M tasks, and highlights the difference between progress and ordering.

A detailed technical article explaining the Linux kernel scheduler, including task_struct, scheduling classes, context switching, and the EEVDF algorithm.

Redox OS, a Unix-like microkernel OS written in Rust, shares its May 2026 updates including the announcement of Summer of Code projects, implementation of EEVDF scheduler, and significant performance improvements in I/O events and filesystem inode caching.

Silk is an open-source cooperative fiber scheduler for Linux with per-CPU scheduler threads, io_uring integration, and topology-aware work-stealing, designed for high concurrency with low overhead.

Developer explores how to abstract GPU workloads so they can run across multiple GPU providers without provider-specific configuration, leaning toward separating workload definition from infrastructure binding.

ByteDance has open-sourced Gödel, a high-performance Kubernetes scheduler, contributing it to the open-source community.