Modern scientific discoveries rely on an insatiable demand for computational resources. To meet this ever-growing computing demand, the datacenters have been established, which are complex controlled environments that host thousands of computing nodes, storage, high-performance communication networks, cooling systems, etc. A datacenter consumes a large amount of electrical power (in the range of megawatts), which gets completely transformed into heat, creating complex spatial and temporal thermal dissipation problems. Therefore, although a datacenter contains sophisticated cooling systems, minor thermal issues/anomalies can potentially trigger a chain of events that leads to an imbalance between the heat generated by computing nodes and the heat removed by the cooling system, leading to thermal hazards. Thermal hazards are detrimental to datacenter operations as they can lead to IT and facility equipment damage as well as an outage of the datacenter, with severe societal and business losses. So, predicting the thermal hazard/anomaly is critical to prevent future disasters. In doing so, collecting and analyzing large-scale monitoring signals and methodology for anomaly detection and prediction are challenging tasks. In this manuscript, after providing a methodology for defining the thermal anomaly, we proposed HazardNet, a thermal hazard prediction framework that consists of a complete pipeline of deep learning models. We evaluated the proposed framework in two different scenarios. In the first scenario, we evaluated the model’s performance over the entire study period, resulting in an F1-score of 0.98. In the second scenario, we enforced causality in the collected data by training and testing the model in two disjunct and consecutive periods, resulting in an F1-score of 0.87. Thanks to these promising results, HazardNet can capture the complex spatial and temporal dependency between datacenter operational parameters and thermal hazards and predict them in advance.
Seyedkazemi Ardebili, M., Acquaviva, A., Benini, L., Bartolini, A. (2024). HazardNet: A thermal hazard prediction framework for datacenters. FUTURE GENERATION COMPUTER SYSTEMS, 155, 340-353 [10.1016/j.future.2024.01.031].
HazardNet: A thermal hazard prediction framework for datacenters
Seyedkazemi Ardebili, Mohsen
Primo
;Acquaviva, Andrea;Benini, Luca;Bartolini, Andrea
Ultimo
2024
Abstract
Modern scientific discoveries rely on an insatiable demand for computational resources. To meet this ever-growing computing demand, the datacenters have been established, which are complex controlled environments that host thousands of computing nodes, storage, high-performance communication networks, cooling systems, etc. A datacenter consumes a large amount of electrical power (in the range of megawatts), which gets completely transformed into heat, creating complex spatial and temporal thermal dissipation problems. Therefore, although a datacenter contains sophisticated cooling systems, minor thermal issues/anomalies can potentially trigger a chain of events that leads to an imbalance between the heat generated by computing nodes and the heat removed by the cooling system, leading to thermal hazards. Thermal hazards are detrimental to datacenter operations as they can lead to IT and facility equipment damage as well as an outage of the datacenter, with severe societal and business losses. So, predicting the thermal hazard/anomaly is critical to prevent future disasters. In doing so, collecting and analyzing large-scale monitoring signals and methodology for anomaly detection and prediction are challenging tasks. In this manuscript, after providing a methodology for defining the thermal anomaly, we proposed HazardNet, a thermal hazard prediction framework that consists of a complete pipeline of deep learning models. We evaluated the proposed framework in two different scenarios. In the first scenario, we evaluated the model’s performance over the entire study period, resulting in an F1-score of 0.98. In the second scenario, we enforced causality in the collected data by training and testing the model in two disjunct and consecutive periods, resulting in an F1-score of 0.87. Thanks to these promising results, HazardNet can capture the complex spatial and temporal dependency between datacenter operational parameters and thermal hazards and predict them in advance.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
