Energy cost minimization for sustainable cloud computing using option pricing

• Published in: Sustainable cities and society 2020-12, Vol.63, p.102440, Article 102440
• Main Authors: Khalil, Muhammad Imran Khan, Ahmad, Iftikhar, Shah, Syed Adeel Ali, Jan, Sadeeq, Khan, Fazal Qudus
• Format: Article
• Language: English
• Subjects: Data centers; Energy efficiency; Geographical load balancing; Option pricing; Sustainable cities; Sustainable cloud computing
• ISSN: 2210-6707; 2210-6715
• DOI: 10.1016/j.scs.2020.102440

One of the key requirements of a sustainable city is the efficiency of the underlying technologies implemented for the provision of services to its residents. Cloud Computing is one such technology that offers a variety of applications support for sustainable cities such as infrastructure for processing of big data and provision of real time services to end users. However, power management of cloud computing infrastructure (data centers) becomes a challenging issue for Cloud Service Providers (CSPs), having negative financial, environmental and sustainability implications. To process the user requests (incoming workload), the data centers (DCs) incur huge energy (electricity) bills. Current studies indicate the workload allocation strategies among geographical distributed DCs mostly focus on using renewable energy and cheaper rates of electricity to reduce the overall energy costs. In this paper, we explore the energy cost minimization problem for geo-distributed DCs considering call option in electricity derivative market under time-varying system dynamics. A call option is an agreement between CSPs and electricity supplier, which gives the rights (not obligation) to option holder to purchase a specific amount of electricity over a time period at a predefined fixed price. To minimize the energy cost, we propose an online algorithm for interactive workload distribution called option pricing based geographical load balancing (OptionGLB). Experimental evaluation based on real-world data indicates the efficacy of OptionGLB over current workload allocation strategies. •Formulating energy minimization problem in data centers considering option pricing.•Determining value of call option using Black–Scholes Model.•Developing workload distribution algorithm using various decision parameters.•Detailed experiments on real world data to assess effectiveness of proposed algorithm.