oom

A sophisticated low memory handler for Linux

根据 Crash/ANR/OOM 等日志构建证据链，实现自动修复代码的Agent框架，并支持衔接自动测试验证、Bug管理平台 Agent framework that builds evidence chains from Crash/ANR/OOM logs, auto-fixes code, and plugs into test verification & bug trackers.

Out Of Memory Analyzer can be used to obtain and summarize "out of memory" issues logged by a Linux kernel (when oom-killer is invoked)

Two environment variables that fix PyTorch/glibc memory creep on Linux. Zero code changes. Zero performance cost.

Depict GPU memory footprint during DNN training of PyTorch

A libray/program that gives information about oom scores

Python and bash scripts to analyze all system process oom_scores and sort by the highest value (most likely to get killed)

ViBrow — кастомный Python-браузер с собственным DOM, CSS и JS движками. Имеет многовкладвочный интерфейс, локальный кэш, cookies, встроенный поисковик и прокси-мост для обхода блокировок. Был создан 1 разработчиком, без сторонних библиотек. Потребляет 30-65МБ ОЗУ, нужен 700МГц процессор 1 ядро и 200МБ ОЗУ LPDDR1. Работает на чистом Python

Icinga2/Nagios check for Out of memory problems

OOM score daemon

System-wide OOM killer for macOS — kills the biggest runaway process before the kernel panic-reboots from swap exhaustion (the earlyoom/systemd-oomd macOS is missing).

Compress AI models to reduce VRAM usage and file size for local LLM deployment and efficient storage.

Prevention, diagnosis and forensics for NVIDIA GB10 AI workstations (DGX Spark, ASUS Ascent GX10, Gigabyte AI TOP ATOM): Xid, thermal, OOM, stuck-GPU-clock detector, per-cgroup GPU memory (dmem patch), guardrails

Command-line Python3 programs to configure/control the Linux Out-Of-Memory (OOM) Killer.

Workload-agnostic Kubernetes OOM watchdog sidecar: rescues memory-bursty containers via in-place pod resize (pods/resize), borrows memory under host-safety guards and returns it when pressure clears.

Cloud-native Kubernetes optimization engine that analyzes workload resource utilization, detects overprovisioned deployments, and generates intelligent CPU and memory recommendations to improve cluster efficiency, scalability, and infrastructure cost optimization using Python, FastAPI, Docker, and Kubernetes systems engineering concepts.

Pre-dispatch RAM/disk/thermal gate for Playwright-class tools next to a local LLM; missing sensors are not 0% free.

macOS AI agent resource hog watcher that sacrifices expendable GUI apps before OOM kills agent work

Send details to Alertmanager when an Out-Of-Memory (OOM) event occurs.

PSI-driven OOM defense for memory-capped Linux VMs — victim steering, controlled sacrifices, and a crash ledger that doubles as experiment data