Cost- and performance-efficient edge analytics is an important research area that aims to deliver real-time data processing and intelligent decision-making at the network edge while minimizing operational costs and optimizing system performance. Research papers in this domain investigate strategies to balance computation, storage, and communication resources for latency-sensitive and resource-intensive applications such as autonomous vehicles, industrial IoT, smart healthcare, and immersive media. Studies explore adaptive workload offloading, dynamic resource allocation, and service orchestration frameworks that leverage optimization algorithms, machine learning, and reinforcement learning to maximize performance while reducing energy consumption and infrastructure costs. Recent works also focus on multi-tier edge–fog–cloud architectures to achieve scalable, fault-tolerant, and cost-aware analytics, incorporating QoS (Quality of Service) and QoE (Quality of Experience) metrics into decision-making. Security- and privacy-preserving approaches, including federated learning and blockchain integration, are increasingly incorporated to ensure data protection without compromising efficiency. Overall, cost- and performance-efficient edge analytics research emphasizes intelligent, adaptive, and resource-aware frameworks that deliver high-quality analytics under constrained budgets and dynamic operational conditions.