Capacity Management of Hyperscale Data Centers Using Predictive Modelling

Abstract

Big Data applications have become increasingly popular with the emergence of cloud computing and the explosion of artificial intelligence. The increasing adoption of data-intensive machines and services is driving the need for more power to keep the data centers of the world running. It has become crucial for large IT companies to monitor the energy efficiency of their data-center facilities and to take actions on the optimization of these heavy electricity consumers. This paper proposes a Belief Rule-Based Expert System (BRBES)-based predictive model to predict the Power Usage Effectiveness (PUE) of a data center. The uniqueness of this model consists of the integration of a novel learning mechanism consisting of parameter and structure optimization by using BRBES-based adaptive Differential Evolution (BRBaDE), significantly improving the accuracy of PUE prediction. This model has been evaluated by using real-world data collected from a Facebook data center located in Luleå, Sweden. In addition, to prove the robustness of the predictive model, it has been compared with other machine learning techniques, such as an Artificial Neural Network (ANN) and an Adaptive Neuro Fuzzy Inference System (ANFIS), where it showed a better result. Further, due to the flexibility of the BRBES-based predictive model, it can be used to capture the nonlinear dependencies of many variables of a data center, allowing the prediction of PUE with much accuracy. Consequently, this plays an important role to make data centers more energy-efficient.

A BRBES to Support Diagnosis of COVID-19 Using Clinical and CT Scan Data

“…These methods rely on ML or DL algorithms whose limitations we discussed in previous section. Unlike ML and DL algorithms, BRBES can address nonlinear causal data, reason under uncertainity and can be optimized using gradient-free nature-inspired algorithms such as BRBaDE [18]. BRBES has been applied to design artificial intelligence-based diagnosis systems for various diseases and medical conditions [32][33][34][35][36][37][38][39].…”

“…However, the crossover and mutation factors remain constant hindering balanced exploration and exploitation of the search space. This has been addressed in [18] where a new optimization algorithm, BRBES-based adaptive differential evolution (BRBaDE) has been used to optimize a BRBES for power usage effectiveness (PUE) prediction of data centers. In summary, BRBES can infer on hematological and chest CT scan data of COVID-19 patients that contain uncertainty and BRBaDE provides a gradient free optimization with optimal exploration and exploitation.…”

“…The objective function ξ is constructed using Eqs. 7 and 8 of [18]. The optimization of the parameters is done using the following: min P ξ(P)…”

“…Also, BRBES can learn using gradient-free optimization approaches such as various evolutionary algorithms. BRBaDE [18] is one of the efficient evolutionary algorithms that provides an appropriate balance between exploration and exploitation of the search space.…”

Artificial Intelligence Research and Its Contributions to the European Union’s Political Governance: Comparative Study between Member States

“…In the fields of business, management and accounting (Figure 3) with regard to the use of AI, Sweden has been an example of success, as was the use of energy management of hyper-scale data centers using predictive modeling (Islam et al 2019). In Islam et al's (2019) research, real data were used from a Facebook data center located in Luleå, Sweden, which allowed managing energy for better management through machine learning techniques, such as an artificial neural network (ANN) and a system adaptive inference neuro fuzzy (ANFIS).…”