ExaDigiT

ExaDigiT/RAPS is a simulator for replaying or modeling workloads on supercomputers and predicting the resulting power consumption.

HTML source for the ExaDigiT project landing page (exadigit.github.io)

ExaDigiT apps and extensions for NVIDIA Omniverse, with support for Apple Vision Pro (AVP).

ExaDigiT/exadigitUE5 is an UnrealEngine5 based Augmented Reality Visual Analytics tool for modeling and visualizing supercomputers as a digital twin; it includes real-time interaction with telemetry and simulations, which are part of the larger ExaDigiT project.

AutoCSM-based approach for modeling cooling systems for POWER9 systems, e.g., Summit, Marconi100, Lassen.

An Unreal Engine plugin for visualizing and interacting with hierarchical datacenter infrastructure through JSON configurations.

ExaDigiT/datacenterCoolingModel is a Modelica-based thermo-fluid cooling model based on Frontier's cooling system built in the commercial IDE Dymola using open-source libraries, which can predict the cooling system performance and its power consumption.

REST API interface and server for RAPS simulations.

Web-based dashboard interface to RAPS. Makes REST API calls to SimulationServer.

ExaDigiT documentation in Python Sphinx.