Current cloud computing frameworks host millions of physical servers that utilize cloud computing resources in the form of different virtual machines. Cloud Data Center (CDC) infrastructures require significant amounts of energy to deliver large scale computational services. Moreover, computing nodes generate large volumes of heat, requiring cooling units in turn to eliminate the effect of this heat. Thus, overall energy consumption of the CDC increases tremendously for servers as well as for cooling units. However, current workload allocation policies do not take into account effect on temperature and it is challenging to simulate the thermal behaviour of CDCs. There is a need for a thermal-aware framework to simulate and model the behaviour of nodes and measure the important performance parameters which can be affected by its temperature. In this paper, we propose a lightweight framework, ThermoSim, for modelling and simulation of thermal-aware resource management for cloud computing environments. This work presents a Recurrent Neural Network based deep learning temperature predictor for CDCs which is utilized by ThermoSim for lightweight resource management in constrained cloud environments. ThermoSim extends the CloudSim toolkit helping to analyse the performance of various key parameters such as energy consumption, service level agreement violation rate, number of virtual machine migrations and temperature during the management of cloud resources for execution of workloads. Further, different energy-aware and thermal-aware resource management techniques are tested using the proposed ThermoSim framework in order to validate it against the existing framework (Thas). The experimental results demonstrate the proposed framework is capable of modelling and simulating the thermal behaviour of a CDC and ThermoSim framework is better than Thas in terms of energy consumption, cost, time, memory usage and prediction accuracy.

Introduction

Resource management is critical in cloud environment in which resource utilization, power consumption of servers, and storage play important roles. Provisioning and scheduling cloud resources is often based on availability, without considering other crucial parameters such as resource utilization or the server‘s thermal characteristics [1]. To realize this, a thermal-aware simulator for resource allocation mechanism is required [2]. The problem of allocating user workloads to a set of Physical Machines (PMs) or Virtual Machines (VMs) and allocating VMs on different server farms adhering to the terms of service as cited in Service Level Agreements (SLAs) and sustaining the Quality of Service (QoS) is stated as the service provisioning issue. Thus, cloud providers focus on developing energy-efficient approaches and policies [4]. Thermo-awareness in cloud refers to the consideration of thermal properties, such as thermal temperature of the host, CPU temperature, heat tolerance and thresholds, energy source (i.e. non-renewable vs. renewable), cooling considerations and mechanisms, cost etc. when dynamically managing cloud resources, scheduling and allocating workloads [20]. The explicit consideration of these properties can transform the way the cloud is managed and resources/PMs/VMs are dynamically allocated, leading to more energy-efficient computing and reduced carbon footprint [24] [35].