Power supply of big infrastructures is today a tremendous operational cost for providers and the expected growth of Internet traffic and services will lead to a further expansion of the computing and networking infrastructures and this, in its turn, raises also concerns in terms of sustainability. In this context, renewable energy generators can help to both reduce costs and alleviate the concerns of sustainability of big infrastructures. In this paper, we consider the case of Data Centers (DCs) composed of a few sites located in different geographical positions and powered with solar energy. Due to the intermittent nature of solar energy, different time zones and price of electricity in different locations, load management strategies are fundamental. We consider predictions of the solar energy production performed through Artificial Neural Networks and we assess the impact of predictions on load management decisions and, ultimately, on the DC performance.
Load Management with Predictions of Solar Energy Production for Cloud Data Centers / Floridia, M.; Lagana, D.; Mastroianni, C.; Meo, M.; Renga, D.. - (2020), pp. 8986-8990. (Intervento presentato al convegno 2020 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2020 tenutosi a esp nel 2020) [10.1109/ICASSP40776.2020.9053186].
Load Management with Predictions of Solar Energy Production for Cloud Data Centers
Meo M.;Renga D.
2020
Abstract
Power supply of big infrastructures is today a tremendous operational cost for providers and the expected growth of Internet traffic and services will lead to a further expansion of the computing and networking infrastructures and this, in its turn, raises also concerns in terms of sustainability. In this context, renewable energy generators can help to both reduce costs and alleviate the concerns of sustainability of big infrastructures. In this paper, we consider the case of Data Centers (DCs) composed of a few sites located in different geographical positions and powered with solar energy. Due to the intermittent nature of solar energy, different time zones and price of electricity in different locations, load management strategies are fundamental. We consider predictions of the solar energy production performed through Artificial Neural Networks and we assess the impact of predictions on load management decisions and, ultimately, on the DC performance.File | Dimensione | Formato | |
---|---|---|---|
2019_ICASSP_datacenter(1).pdf
accesso aperto
Tipologia:
2. Post-print / Author's Accepted Manuscript
Licenza:
Pubblico - Tutti i diritti riservati
Dimensione
448.27 kB
Formato
Adobe PDF
|
448.27 kB | Adobe PDF | Visualizza/Apri |
Renga-Load.pdf
accesso riservato
Tipologia:
2a Post-print versione editoriale / Version of Record
Licenza:
Non Pubblico - Accesso privato/ristretto
Dimensione
441.86 kB
Formato
Adobe PDF
|
441.86 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/11583/2853817