Providing Energy-efficient and High-performance Infrastructure for Smart Network

Energy efficiency, performance and service availability are the key performance indicators of modern Smart Networks. These networks use large-scale distributed computing systems to provide high-performance data processing. In such systems, it is always difficult to find a balance between performance and power consumption, given the possible heterogeneity of nodes in a distributed system. Existing approaches propose using horizontal scaling, workload consolidation, and energy-efficient scheduling to improve the energy efficiency of large-scale distributed systems while maintaining sufficient processing performance. We propose to combine the benefits of horizontal scaling and energy efficient scheduling and model the system as a queuing system to achieve the optimal balance of performance, power consumption, and service availability in our PCPB-2 approach. Experimental analysis of the approach proves its efficiency up to 8.9% compared to the state-of-the-art Backfill energy-aware scheduling approach.