Adaptive Task Scheduling in Urban Vehicular Cloud Networks
College
College of Sciences
Department
Computer Science
Graduate Level
Doctoral
Presentation Type
Poster Presentation
Abstract
Vehicular cloud computing in urban environments requires a flexible and adaptive approach to resource allocation due to the dynamic nature of vehicle availability. The Flexible Time Datacenter (FTDC) framework extends beyond smart parking lots, utilizing C-V2X communication to integrate both moving and stationary vehicles into a computational network. This research work introduces FTDC, which employs a truthful reverse auction mechanism to efficiently allocate computational resources among vehicles and tasks. By dynamically matching tasks with available vehicles, the framework optimizes resource utilization while accounting for disruptions such as vehicles entering maintenance mode or unexpectedly leaving the network. To address these challenges, FTDC includes a task mitigation strategy that swiftly reallocates or re-auctions interrupted tasks, ensuring deadline adherence and service continuity. The system also incorporates incentives for reliable participation and penalties for early withdrawals to maintain operational efficiency. Through extensive simulations, the FTDC framework is evaluated for its effectiveness in handling task allocation, optimizing resource use, and ensuring system reliability. The results demonstrate its adaptability to the complexities of urban vehicular networks while minimizing delays and maximizing performance.
Keywords
Vehicular Cloud Computing, Dynamic Resource Allocation, C-V2X Communication, Task Reallocation
Adaptive Task Scheduling in Urban Vehicular Cloud Networks
Vehicular cloud computing in urban environments requires a flexible and adaptive approach to resource allocation due to the dynamic nature of vehicle availability. The Flexible Time Datacenter (FTDC) framework extends beyond smart parking lots, utilizing C-V2X communication to integrate both moving and stationary vehicles into a computational network. This research work introduces FTDC, which employs a truthful reverse auction mechanism to efficiently allocate computational resources among vehicles and tasks. By dynamically matching tasks with available vehicles, the framework optimizes resource utilization while accounting for disruptions such as vehicles entering maintenance mode or unexpectedly leaving the network. To address these challenges, FTDC includes a task mitigation strategy that swiftly reallocates or re-auctions interrupted tasks, ensuring deadline adherence and service continuity. The system also incorporates incentives for reliable participation and penalties for early withdrawals to maintain operational efficiency. Through extensive simulations, the FTDC framework is evaluated for its effectiveness in handling task allocation, optimizing resource use, and ensuring system reliability. The results demonstrate its adaptability to the complexities of urban vehicular networks while minimizing delays and maximizing performance.