Abstract
Recent advances in computing and sensing techniques lead to the generation of tons of data that needs to be processed in distributed computing infrastructures (e.g., datacenters). In this paper, we focus on applying optical networking techniques for supporting the big data transfer between datacenters. In particular, we propose a green optical networking scheme, namely predictive and incremental grooming (PI-grooming), which makes the traffic grooming decision based on the existing flow assignment, the current traffic demands, and the expected future traffic demands to achieve energy-efficient communication. We conduct large-scale simulations to evaluate the performance of the proposed PI-grooming scheme. The results show that PI-grooming can save energy by up to 25% compared to existing baseline solutions.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
