SINC: Scheduling Interdependent Network Chains of Coalesced Flows and Functions
Abstract
To enable a vision toward future networks with full autonomy, network management is quickly shiftingfrom packet/flow based models that mainly focus on traffic engineering to more complex problemsof jointly optimizing a collection of closely-interacting traffic flows and network functions, which areconsidered as Service Function Chains (SFCs). An SFC is defined as an ordered sequence of network functions and corresponding data flows traveling through them to provide end-to-end services and capabilities. An SFC scheduling problem, which aims to optimally map network functions and relevant data flows in a set of SFCs onto server nodes and links in underlying physical network, respectively, is becoming increasingly important.To this end, the proposed research will support an efficientand flexible network management architecture for many maritime tasks. It will investigate a novel SFC scheduling framework under precedence and deadline constraints and quantify the latency in dynamic SFC scheduling problems with stochastic SFC arrivals and random service times. Specifically, we will model precedence constraints as general acyclic graphs, study SFC scheduling under such constraints, and propose a queuing model for characterizing the latency. To enable more scalable solutions, we will also investigate a learning-based approach to solve SFC scheduling through sequential decision making, potentially with partial and incomplete observations, thus making our solutions more practical. The project will enable an operational architecture for distributed maritime operations, connecting distributed units into groups and distributed groups into fleets. At the core of this architecture is the tactical grid, in which different network functions and related network flows # modeled as SFCs in this project # must work in concert to provide timely, seamless, and robust connectivity under dynamic (and even hostile) environments and to achieve critical objectives.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- May 15, 2023
- Source ID
- N000142312532
Entities
People
- Tian Lan
Organizations
- George Washington University
- Office of Naval Research
- United States Navy