Towards designing a green data center farm for Internet services: Iran’s case study

Abstract

Energy consumption has become a critical design factor in today’s data centers. In recent years, extensive research has been done to address power–performance trade-off in data centers considering both IT equipments and cooling infrastructures (e.g., thermal-aware task scheduling, server consolidation, load balancing and geographical load balancing to name a few). This paper introduces a design-time technique that targets energy-efficient design of a green data center farm in Iran. Workload predictions, geographical maps of wind speed and solar radiation, data center and renewable resources configurations are used as a priori to design an energy-efficient data center farm for Internet services. The proposed problem is mathematically formulated as a nonlinear optimization problem and is effectively solved using a coordinate descent-based method. We also show that with some minor modification, our proposed technique can be applied at run-time for the purpose of change management. The experimental results show that the proposed method can lead to 11.6 % cost saving on average over conventional approaches.

Appendix: Parametrization coefficients

To calculate the cost function in Phase-I of our proposed algorithm (data center right-sizing/workload scheduling step), we define $\omega _j$, $\alpha _{j}$, $\beta _{j}$, $\sigma _{ij}$, $\theta _{ij}$ and $\rho _{ij}$ as a function of other fixed parameters as follows:

$$\begin{aligned}&\omega _j = \frac{P_{\mathrm{sla}}^j}{C_{\text {IT}}/R_{\text {inf}}+C_eTP_{\text {IT}}+C_\text {MT}} \end{aligned}$$

(40)

$$\begin{aligned}&\alpha _{j} = -\mu \tau ^{\text {max}}_j \end{aligned}$$

(41)

$$\begin{aligned}&\beta _{j} = \tau ^{\text {max}}_j \end{aligned}$$

(42)

$$\begin{aligned}&\sigma _{ij} = 1+\mu (\tau _d^{ij}+\tau _0) \end{aligned}$$

(43)

$$\begin{aligned}&\theta _{ij} = -(\tau _d^{ij}+\tau _0) \end{aligned}$$

(44)

$$\begin{aligned}&\rho _{ij} = \frac{1}{\mu +\frac{1}{\frac{\tau ^{\text {max}}_j}{ln(h_j)}+\tau _d^{ij}+\tau _0}}. \end{aligned}$$

(45)