CF'26, Catania, Italy · May 2026
Datacenters are vital to our digital society, but consume a considerable fraction of global electricity and demand is projected to increase. To improve their sustainability and performance, we envision that simulators will become primary decision-making tools. However, and unlike other fields focusing on key societal infrastructure such as waterworks and mass transit, datacenter simulators do not yet combine multiple independent models into their operation...
Abstract
Talk
Datacenters are vital to our digital society, but consume a considerable fraction of global electricity and demand is projected to increase. To improve their sustainability and performance, we envision that simulators will become primary decision-making tools. However, and unlike other fields focusing on key societal infrastructure such as waterworks and mass transit, datacenter simulators do not yet combine multiple independent models into their operation and thus suffer from issues associated with singular models, such as specialization, and lack of adaptability to operational phenomena. To address this challenge, we propose M3SA, a datacenter simulation and analysis framework that uses discrete-event simulation to predict, for each model, the impact on climate and performance under various realistic datacenter conditions, and then combines these predictions. We design an architecture for simulating multiple concurrent models (Multi-Model), a technique for integrating the results of multiple models into a Meta-Model, and a procedure for quantifying Meta-Model accuracy. Through experiments with an M3SA prototype, we reproduce and enhance a peer-reviewed experiment with multi-model analysis and identify that M3SA can halve error rates of singular models (from 7.59% to 3.81%), with under 20% computational overhead.
ICPE'26, Florence, Italy · May 2026
Datacenters are the backbone of our digital society, but raise numerous operational challenges. We envision digital twins becoming primary instruments in datacenter operations, continuously and autonomously helping with major operational decisions and with adapting ICT infrastructure, live, with a human-in-the-loop. Although fields such as aviation...
Abstract
Talk
Datacenters are the backbone of our digital society, but raise numerous operational challenges. We envision digital twins becoming primary instruments in datacenter operations, continuously and autonomously helping with major operational decisions and with adapting ICT infrastructure, live, with a human-in-the-loop. Although fields such as aviation and autonomous driving successfully employ digital twins, an open-source digital twin for datacenters has not been demonstrated to the community. Addressing this challenge, we design, implement, and experiment using OpenDT, an Open-source, Digital Twin for monitoring and operating datacenters through a continuous integration cycle that includes: (1) live and continuous telemetry data; (2) discrete-event simulation using live telemetry from the physical ICT, with self-calibration; and (3) SLO-aware and human-approved feedback to physical ICT. Through trace-driven experiments with a prototype mainly covering stages 1 and 2 of the cycle, we show that (i) OpenDT can be used to reproduce peer-reviewed experiments and extend the analysis with performance and energy-efficiency results; (ii) OpenDT's online re-calibration can increase digital-twinning accuracy, quantified to a MAPE of 4.39% vs. 7.86% in peer-reviewed work. OpenDT adheres to FAIR/FOSS principles and is available at: https://github.com/atlarge-research/opendt/tree/hcp.
Winter School on Generative AI in Academia, VU Amsterdam · January 2026
Lecture 4 of the Winter School on the Use of Generative AI in Academia covers Gen AI infrastructure and datacenters. LLMs run in big computers—datacenters—and this is the golden age of datacenters: they are crucial to today's society, widely used worldwide amongst academia, government, and industry. Yet we cannot take this technology for granted, particularly when considering economic and climate sustainability of massive computer ecosystems...
Abstract
Talk
Lecture 4 of the Winter School on the Use of Generative AI in Academia covers Gen AI infrastructure and datacenters. LLMs run in big computers—datacenters—and this is the golden age of datacenters: they are crucial to today's society, widely used worldwide amongst academia, government, and industry. Yet we cannot take this technology for granted, particularly when considering economic and climate sustainability of massive computer ecosystems. The lecture explains how GenAI runs on the compute continuum—from endpoint (e.g., phone) to edge (e.g., midway DC) to cloud (e.g., large DC)— following a reference architecture for LLM ecosystems under inference. GenAI is a large iceberg; there is a lot happening "under the hood", and GenAI is largely distributed over the compute continuum. To understand and improve GenAI IT operations, the talk covers simulating datacenters, simulating LLM inference, and digital twinning ICT and datacenters. Real-world experimentation is costly; simulation allows cheap, fast, accessible what-if analysis. The lecture contrasts simulation (one time) with digital twinning (continuous), presents work on datacenter and LLM inference simulation, and introduces OpenDT, the first open-source digital twin for operating and monitoring datacenters—addressing the gap that, until now, there has been no digital twin for ICT.