Custom Spawner for Jupyterhub to start servers in batch scheduled systems
-
Updated
Oct 5, 2026 - Python
Custom Spawner for Jupyterhub to start servers in batch scheduled systems
Longbow is a tool for automating simulations on a remote HPC machine. Longbow is designed to mimic the normal way an application is run locally but allows simulations to be sent to powerful machines.
LLM + RAG system for querying HPC platform documentation and codebase — document crawling, chunking, embeddings, vector search, and LangChain-based retrieval-augmented generation.
Hands-on guide to training and fine-tuning LLMs on HPC clusters
Parallel 2D Laplace solver implemented using MPI and OpenMP on NUST RCMS HPC. Benchmarks distributed vs shared memory parallelism on AMD EPYC 7452 with performance graphs and correctness verification.
LaserFire module. part of the K-9 and supercomputer projects
HPC System Description Schema
Deep learning framework for thermal hazard (anomaly) prediction in HPC datacenters and supercomputers. TCN, LSTM, and SVM time-series models forecast forthcoming thermal hazards from compute-node temperature and power sensors (Marconi A2 / CINECA). Python, PyTorch, TensorFlow.
Parallel Python tool to extract and back up Examon HPC monitoring telemetry from the Marconi100 supercomputer (CINECA). Uses ExamonQL over KairosDB to download per-metric, per-plugin time-series (Ganglia, IPMI, Slurm, Nagios, Vertiv) as daily gzipped CSVs, with incremental follow-file tracking, gap back-filling, and multiprocessing.
Distributed YOLOv8 red vehicle detection cluster using Firebase Realtime Database with a live web dashboard simulating a supercomputer.
A flexible package manager designed to support multiple versions, configurations, platforms, and compilers.
To associate your repository with the supercomputer topic, visit your repo's landing page and select "manage topics."