Applications of scheduling theory to optimize green energy usage in cloud computing platforms

• Grant number: 21/06867-2
• Support Opportunities: Regular Research Grants
• Start date: January 01, 2022
• End date: December 31, 2023
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computer Systems
• Agreement: MCTI/MC
• Principal Investigator: Daniel de Angelis Cordeiro
• Grantee: Daniel de Angelis Cordeiro
• Host Institution: Escola de Artes, Ciências e Humanidades (EACH). Universidade de São Paulo (USP). São Paulo , SP, Brazil
• Associated researchers: Emilio de Camargo Francesquini ; Fanny Dufossé

Abstract

Cloud computing platforms support most services and applications that we use every day, such as social networks, e-mail, video games, and video streaming, and is a key element for the development of smart cities. However, cloud computing platforms consume a massive amount of electricity: the data centers that host these platforms consume 1% of the power generated globally. In order to reduce the costs and environmental impact resulting from this energy consumption, data centers are installing renewable energy sources such as solar farms in their facilities. The availability of solar energy is not constant, resulting in challenges in scheduling tasks to reduce nonrenewable energy consumption. Some data centers have batteries that can store renewable energy, but they self-discharge and lose their capacity as time goes on, which characterizes another challenge: deciding when to store or use battery power. In this project, we will study how to use scheduling theory to reduce brown energy consumption and costs for data center operators. In order to tackle the scientific and technological challenges to solve this problem, we will develop a multi-objective scheduling algorithm for virtual machines that are submitted to geographically distributed cloud computing platforms, considering that they have batteries and variable (intermittent) renewable energy supply such as energy from wind and solar farms. We will also propose a method to estimate the dimensions of the renewable power source and batteries, given the size of data centers and their expected workloads. (AU)

Scientific publications

(4)

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

AMARIS, MARCOS; CAMARGO, RAPHAEL; CORDEIRO, DANIEL; GOLDMAN, ALFREDO; TRYSTRAM, DENIS. Evaluating execution time predictions on GPU kernels using an analytical model and machine learning techniques. JOURNAL OF PARALLEL AND DISTRIBUTED COMPUTING, v. 171, p. 13-pg., 2023-01-01. (19/26702-8, 15/19399-6, 21/06867-2, 12/23300-7)

SILVA VASCONCELOS, MIGUEL FELIPE; CORDEIRO, DANIEL; DUFOSSE, FANNY; KLEIN, C; JARKE, M. Indirect Network Impact on the Energy Consumption in Multi-clouds for Follow-the-renewables Approaches. PROCEEDINGS OF THE 11TH INTERNATIONAL CONFERENCE ON SMART CITIES AND GREEN ICT SYSTEMS (SMARTGREENS), v. N/A, p. 12-pg., 2022-01-01. (15/24485-9, 21/06867-2, 14/50937-1)

VASCONCELOS, MIGUEL; CORDEIRO, DANIEL; DA COSTA, GEORGES; DUFOSSE, FANNY; NICOD, JEAN-MARC; REHN-SONIGO, VERONIKA; ALTINTAS, I; SIMMHAN, Y; VARBANESCU, AL; BALAJI, P; et al. Optimal sizing of a globally distributed low carbon cloud federation. 2023 IEEE/ACM 23RD INTERNATIONAL SYMPOSIUM ON CLUSTER, CLOUD AND INTERNET COMPUTING, CCGRID, v. N/A, p. 13-pg., 2023-01-01. (21/06867-2)

FERNANDES, MATHEUS CAMPOS; DE FRANCA, FABRICIO OLIVETTI; FRANCESQUINI, EMILIO; PAQUETE, L. HOTGP- Higher-Order Typed Genetic Programming. PROCEEDINGS OF THE 2023 GENETIC AND EVOLUTIONARY COMPUTATION CONFERENCE, GECCO 2023, v. N/A, p. 9-pg., 2023-01-01. (19/26702-8, 21/06867-2, 21/12706-1)