Publications

Geo-distributed data centres (DCs) that recently established due to the increasing use of on-demand cloud services have increasingly attracted cloud providers as well as researchers attention. Energy and data transmission cost are two significant problems that degrades the cloud provider net profit. However, increasing awareness about CO2 emissions leads to a greater demand for cleaner products and services. Most of the proposed approaches tackle these problems separately. This paper proposes green approach for joint management of virtual machine (VM) and data placement that results in less energy consumption, less CO2 emission, and less access latency towards large-scale cloud providers operational cost minimization. To advance the performance of the proposed model, a novel machine-learning model was constructed. Extensive simulation using synthetic and real data are conducted using the CloudSim simulator to validate the effectiveness of the proposed model. The promising results approve the efficacy of the CELA model compared to other competing models in reducing network latency, energy consumption, CO2 emission and total cloud provider operational cost.