As microservices and containerized architectures gain prominence, Kubernetes has become the foundation for orchestrating distributed workloads. However, its default scheduling strategy is typically static and resource-centric, lacking responsiveness to changing workload patterns, inter-service communication needs, or energy efficiency goals. This paper introduces AICoLoc-K8s, an adaptive scheduling enhancement framework that augments Kubernetes with intelligent, real-time decision-making. AICoLoc-K8s leverages live system metrics collected from Prometheus and a lightweight AI model to inform pod placement decisions. Integrating a custom scheduler extender, the system dynamically evaluates candidate nodes based on multiple criteria, including CPU load, memory availability, and service tier affinity. The AI model is trained to optimize pod co-location, especially for workloads categorized by frontend, backend, and database roles. We implemented AICoLoc-K8s on an RKE2 cluster running inside KubeVirt virtual machines and validated its behavior through controlled experiments. The framework consistently demonstrated better placement alignment with workload characteristics than the default scheduler. Although quantifiable gains like latency and energy savings are reserved for future evaluation, initial results confirm the effectiveness of this real-time adaptive model.

The Invisible Internet Project (I2P) is a decentralized peer-to-peer anonymity network that protects users’ privacy by routing traffic through encrypted tunnels across volunteer-run routers (nodes). Its distributed nature raises critical questions about structural resilience, specifically, how well it can withstand random (stochastic) failures and targeted (adversarial) attacks. This study models I2P’s overlay using three representative network graphs or topologies: Random Graph (RG), Scale-Free (SF), and a theoretical modeling of I2P’s network, herein referred to as I2P Prime (I2P’), all experimented with 50,000 nodes (peers) each to reflect the real-world conditions of the I2P network. Simulations under random node removals show that all three networks retain large connected components (LCC) beyond 50% node loss, with I2P Prime maintaining superior efficiency. However, targeted attacks based on degree or betweenness centrality reveal substantial vulnerabilities. The SF network model of I2P collapsed rapidly, often below 30% node removal due to its hub-centric design. In contrast, I2P Prime exhibits stronger fault tolerance, requiring nearly 50% of critical nodes to be removed before global connectivity fails. These findings underscore the structural advantages of topologies like I2P Prime, which combines distributed connectivity and resilience to percolation and Perturbation. For developers, enhanced adaptive peer selection and dynamic routing mechanisms could enhance robustness without undermining anonymity. For policymakers, our results highlight how targeted interventions might fragment illicit activity with minimal collateral impact. This work provides actionable insights into designing resilient anonymity networks that preserve privacy under stochastic and adversarial attacks.

Network security is experiencing huge challenges as network attacks on traffic data become more frequent and sophisticated. In this paper, we employ hybrid deep learning models and low-rank approximation to present a novel method for multi-label categorization of network assaults on traffic data. Our suggested solution, LR-CNN-MLP, consists of three models While the CNN and MLP models extract features and categorise data, respectively, the low-rank approximation model reduces the input's dimensionality. Overall, by combining hybrid models and low-rank approximation, our proposed LR-CNN-MLP approach provides a promising solution for multi-label categorization of network attacks on traffic data.

A data center (DC) is supposed to efficiently distribute the bandwidth of the network to provide high-quality traffic transmission. However, the load imbalance issue can easily occur due to the complex topology and traffic features. Equal-Cost Multi-Path(ECMP) distributes traffic on different paths but doesn’t consider network congestion. Although HULA solved some of ECMP’s problems, it can easily congest the best path. RPS randomly distributes packets on paths, which can easily lead to packet disorder in some scenarios. This paper presents DHLB, a distributed hop-by-hop load balancing archetucture based on in-band network telemetry. With active In-band network telemetry, DHLB collects the necessary load information and stores it in the load information table. DHLB distributes traffic proportionally on different paths based on their load degree. We build a fat tree topology on mininet to verify the performance of our design. From experimental results, DHLB performs better than other schemes in terms of average flow complete time(FCT). It also performs better on additional overhead than another probe-based scheme.

Network reliability analysis is vital for ensuring efficient and error-free communication within networking and communication applications. Binary Decision Diagrams (BDDs) have emerged as a powerful tool for analyzing and optimizing complex network infrastructures. The objective of this research paper is to conduct a comparative analysis of edge-ordering algorithms for network reliability using BDDs the study aims to evaluate and compare existing algorithms, providing valuable insights for selecting suitable edge-ordering algorithms that enhance network reliability. The paper concludes that snooker is outperforming among selected algorithms.

In this letter, we consider the privacy-preserving strategies of the distributed state estimator over wireless sensor networks. Aiming at an eavesdropper who can intercept the data transmitted on the communication channel, we design a simple preserving scheme to encode the transmitted data by using system dynamics and history estimates. To analyze the privacy-preserving performance, we use the gap between the values deciphered by the eavesdropper and decoded by the sensor as the privacy index to evaluate the degree of robust privacy protection. Finally, we provide some simulations to illustrate the effectiveness of the proposed preserving schemes.