Connectivity is the essential feature in the modern world and is considered a utility — always on and available everywhere. 5G is the fifth generation of wireless technology that facilitates this connectivity with ultra-low network communication latency. 5G will have a massive impact on how we live, learn, work and play.

Virtual Network Functions (VNF), Network Service and Slice (NS) Management and Orchestration (MANO) are the paramount goals for telecommunication operators to rapidly introduce new 5G features like Enhanced Mobile Broadband (eMBB), Massive Machine Type Communications (MMTC), and Ultra-Reliable and Low Latency Communications (URLLC). Network Function Virtualization (NFV) is the critical enabler to commercialize these 5G features. NFV provides abstraction from the substrate physical hardware and utilizes Virtualized Infrastructure Management’s (VIM) dynamic resource allocation and VNFs deployment to deliver the 5G Network Services and Slicing. However, End-to-End (E2E) VNF, Network Service and

Network Slice latency resulting from default VIM-based resource allocation for VNFs deployment may degrade the VNFs, Network Services and Slices’ network communication latency and performance. VNFs, Network Service and Network Slices’ VIM deployments supported with substrate physical inter and intra-(servers, racks, and data centers) communication links among VNFs incur higher E2E latencies and Service Level Agreements (SLA) violations. These network communication delays cause timeouts and faults with latency-sensitive operations. Conventional solutions to reducing E2E latency involve allocating physical resources and deploying physical network elements manually and nearby. This approach is no longer viable in the NFV era.

IllumiCore addresses the VNFs, Network Services and Slices network communication latency problem by efficiently allocating VIM resources and coordinating the VNFs deployment and composition of Network Service and Slice chains on the VIM layer and corresponding Network Function Virtualization Infrastructure (NFVI) substrate hardware infrastructure and networks with Information Model of mapped physical and virtual resources, objective function, and constraint programming solution. IllumiCore aims to minimize VNFs, Network Service and Slice latency between VNF’s Virtual Machines (VM) or Containers within a reasonable resource allocation time. The experimental results demonstrate optimized VNF placement within Network Service and Network Slice chains leading to overall performance improvements and reducing E2E Network Service and Slice network communication latencies while outperforming existing VIM’s filtering, and default Evenly Distributed, and Worst-Fit placement algorithms.

As the future work, we plan to explore the objective function of minimizing the VNFs and Network Services power draw and power consumption while defining efficient placement, SLA, minimize cost and carbon emission in distributed clouds constraints. While IllumiCore helps to allocate resources for VNF placement efficiently, it could be further improved with predictive

Machine Learning (ML) implementation. In our future work, we are planning to use ML to automatically learn from real VNF, VIM, and NFVI data, deriving models that can accurately predict optimal compute, storage and networking resource requirements for the efficient VNF and Network Services placement and further enhance the reliability of VNF latency modeling.

Degree Date

Spring 5-14-2022

Document Type


Degree Name



Computer Science and Engineering


Suku Nair

Subject Area

Computer Science



Creative Commons License

Creative Commons Attribution-Noncommercial 4.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial 4.0 License