Joint balancing of IT and spectrum resources for selecting virtualized network function in inter-datacenter elastic...

Area of Science:

• Computer Networking
• Telecommunications Engineering
• Cloud Computing

Background:

• Network Function Virtualization (NFV) decouples network functions from hardware, enabling faster service provisioning in data centers (DCs).
• Virtualized Network Functions (VNFs) are often deployed across multiple DCs for survivability in inter-datacenter elastic optical networks.
• VNF selection significantly impacts both IT resources within DCs and spectrum resources in optical networks.

Purpose of the Study:

• To investigate optimal VNF selection for service requests to achieve joint load balancing of IT and spectrum resources.
• To develop metrics for quantifying the impact of VNF selections on network load.
• To design an algorithm for efficient VNF selection in NFV environments.

Main Methods:

• Proposed two joint balancing factors to quantify the impact of VNF selections on network load.
• Designed a Joint-Optimization Selection (JOS) algorithm for VNF selection based on joint load balancing.
• Evaluated algorithm performance using simulations focusing on resource utilization, blocking probability, and average path length.

Main Results:

• The proposed JOS algorithm demonstrates superior performance compared to a benchmark algorithm.
• Achieved improved resource utilization in both IT and spectrum domains.
• Showcased reduced blocking probability and average path length.

Conclusions:

• The JOS algorithm effectively balances IT and spectrum resources in NFV-enabled inter-datacenter networks.
• Joint load balancing through optimized VNF selection is crucial for efficient network operation.
• The proposed approach offers significant improvements in key network performance metrics.