The widespread adoption of Large Language Models (LLMs) in enterprise applications has created a critical need for systematic evaluation of Generative AI Systems (GenAIS) that integrate orchestration frameworks, foundation models, and deployment optimizations. This paper presents the first comprehensive evaluation study specifically designed to evaluate the performance trade-offs between orchestration frameworks (LangChain, LlamaIndex), foundation models, and deployment constraints across multiple application domains. Our study systematically evaluates eight foundation models across question-answering with Retrieval-Augmented Generation (RAG) and mathematical reasoning tasks, measuring latency, accuracy, resource utilization, and power consumption under various optimization strategies including adaptive context windowing and concurrent request processing. Through extensive empirical analysis, we demonstrate that framework selection significantly impacts system performance independent of model choice, with task-specific trade-offs emerging across workload types. For retrieval-heavy RAG workloads, LlamaIndex delivers 3-15% lower latency, around 60% lower peak memory usage, and 2-4% more energy consumption. For reasoning-intensive mathematical tasks, Langchain achieves higher accuracy across all tested models and provides more predictable latency profiles and reasonable resource consumption, making it preferable for strict SLA environments. Adaptive context windowing does not yield uniform gains and offers only modest, model-dependent improvements over fixed policies, whereas increasing concurrency improves aggregate throughput but, beyond moderate loads, consistently inflates tail latency and degrades SLA predictability. These findings underscore the decisive role of orchestration design in GenAIS performance and provide empirical guidance for balancing efficiency, accuracy, and scalability in real-world deployments.

Vehicular Cloud Networks (VCNs) represent a promising paradigm for delivering dynamic, low-latency services by leveraging the resources of connected vehicles. However, the highly mobile and resource-constrained nature of VCNs poses significant challenges for reliable service allocation, particularly when considering user-specific preferences. In this paper, we propose a user-preference-based service allocation mechanism that optimizes three key Service Level Agreement (SLA) parameters: delay, availability, and price. Our approach utilizes the Non-dominated Sorting Genetic Algorithm II (NSGA-II) to generate Pareto-optimal solutions, allowing users to select services that best match their preferences. We evaluate the proposed approach through simulation, comparing its performance against alternative methods in terms of execution time, SLA adherence rate, and service stability. The results demonstrate that our approach not only achieves higher SLA adherence but also maintains efficiency under various vehicular conditions.

Generative Adversarial Networks (GANs) have shown promise in enabling intelligent and privacy-preserving data synthesis within decentralized environments like the Internet of Vehicles (IoV). However, applying GANs in such dynamic, resource-constrained, and latency-sensitive settings poses significant challenges, particularly due to intermittent connectivity, mobility-induced disruptions, and communication overhead. In this paper, we propose a novel collaborative Multi-Discriminator GAN (MD-GAN) approach tailored for IoV, where Road Side Units (RSUs) act as generators and vehicles function as mobile discriminators providing distributed feedback. Unlike traditional centralized or synchronous GAN setups, our proposed approach leverages an early update strategy that allows generators to proceed once a minimum threshold of feedback is received, along with a feedback selection mechanism that considers only discriminators offering improved loss scores. We adopt the Wasserstein GAN (WGAN) formulation to ensure stable convergence under sparse and asynchronous feedback conditions. The proposed approach is implemented through a hybrid simulation architecture combining NS3 and PyTorch to model both networklevel interactions and learning dynamics. Experimental results demonstrate that the proposed approach significantly reduces training latency and network overhead compared to the baselines, while maintaining competitive generator accuracy. Our findings highlight the effectiveness of communication-efficient GAN training in highly dynamic vehicular networks.

Vehicular Ad-hoc Networks (VANETs) and in-vehicle networks offer complementary perspectives on Intelligent Transportation Systems (ITS), enabling communication between vehicles and within individual vehicles, respectively. While VANETs focus on vehicle-to-vehicle communication, the growing demand for dynamic resource sharing and data processing across a fleet of vehicles highlights the need for Vehicular Cloud Networks (VCNs). VCNs, despite their lack of fixed infrastructure and the continuous mobility of vehicles, provide a promising solution for improving resource management and data sharing, making them critical for achieving efficient Service Level Agreements (SLAs) in infrastructure-less environments. This paper addresses these challenges by employing a hierarchical clustering technique and proposing a novel mathematical formulation for resource provisioning in infrastructure-less vehicular clouds. The formulation considers diverse criteria, including provider and requester mobility, data volume, and service delay tolerance, to ensure SLA adherence. A customized genetic algorithm is used to solve the maximization problem, incorporating a grouping mechanism for efficient problem solving. Simulations using the NS2 network simulator and the IBM CPLEX optimization tool validate the feasibility of the proposed approach and demonstrate its superior performance compared to the other methods.