In this paper, we compared the existing routing algorithms in the context of Software Defined Networking (SDN), where a logically centralized controller acquires the global view of the network, ...selects the paths using a routing algorithm, and installs the determined routing rules to the switches. We divided existing routing algorithms (RA) into three categories: RA with static link cost (RA-SLC), RA with dynamic link cost (RA-DLC), and RA with dynamic link cost and minimum interference (RA-DLCMI). We then implemented various routing algorithms from each category using RYU SDN controller and compared their performances on Mininet emulator. For the network state information (NSI) needed for the routing algorithms, we first consider the idealistic case where the accurate NSI is available, as assumed in the literature. However, since this is not possible in practice, we also considered the practical case where the controller periodically collects the NSI with some inaccuracy. In our experiments, we observed that while RA-DLC and RA-DLCMI give similar performance, RA-SLC falls behind of RA-DLC and RA-DLCMI in the number of accepted flows and total throughput. We also showed that the performance of every algorithm is adversely affected from the inaccuracy of NSI, calling for further research on developing effective NSI collection methods.
Researchers have been extensively, but often independently, investigating Wireless Sensor Networks (WSNs), Mobile Robotics, and Teleoperation. Lately, due to the complementary nature of these areas, ...there is a growing trend in integrating them to support various applications that can go beyond passive monitoring and allow us to actually interact with the environment through autonomous and teleoperated robots. Such applications can extensively be used in medicine, science, military, industry, nuclear power stations, underwater, and space explorations. In this paper, we start with a review of the history of WSNs, robotics and mobility in WSNs, and teleoperation. We then introduce a system model integrating these areas as Wireless Sensor and Robot Networks with Teleoperation capabilities (WSRNT). Using this model, we define the problems associated with WSRNT. Finally, we discuss the recent research efforts and open research challenges in merging these fields with respect to major WSRNT tasks and communication requirements.
Internet-of-Things (IoT) devices continuously gather data and send the data to the cloud for further processing. However, with the recent trend of increasing number of IoT devices, billions of ...devices are anticipated to send data to the cloud, eventually impacting performance and cost. To solve this problem, one way is to process data locally inside edge nodes. The edge nodes are closer to the IoT devices and improve the overall performance of the system by distributing cloud tasks in edge devices. Edge devices are generally resource-constrained with limited RAM, CPU, and storage. Container technologies are ideal in edge nodes due to their isolation and being lightweight. A benchmarking scheme for containers on edge devices can help compare container technologies, hardware devices and architectures, and software. However, there is not sufficient research in this direction. So, in this work, we take a step towards developing this benchmark. We explore and evaluate the performance, efficiency, and suitability of different container technologies, including Docker, Podman, and Singularity, in the context of edge computing on ARM-based devices. Our experiments include evaluating computer vision applications that employ Haar Cascades, HOG, CNN with YOLO algorithm, and data science workloads commonly encountered in edge computing scenarios. We devised sets of performance metrics to assess container technology, including waiting time, receiving time, processing time, resource utilization, and throughput. Besides, we investigate how different container technologies optimize resource utilization and compare their efficiency on ARM-based edge devices. Our benchmarking analysis yielded valuable insights into the strengths and limitations of each container technology. Our results reveal that Docker, Podman, and Singularity containers exhibit diverse resource consumption patterns and network efficiency. Docker container has better CPU and RAM utilization for most applications. Docker also boasts the lowest waiting time of approximately 0.9 seconds, comparable to native performance. In terms of processing time, Docker excels in Car detection (0.12 seconds), while Singularity and Podman outperform Docker in Object detection. Notably, native systems exhibit a remarkable improvement over containers (average of 1.2 seconds) in Object detection (0.98 seconds), highlighting the challenges of resource-intensive deep neural network algorithms on edge devices. Based on our findings, we offer practical advice for picking the best container technology for specific use cases in ARM-based edge computing. We also offer a set of benchmarking approaches and metrics that can be used to drive future research on container technologies on ARM platforms.
In the dynamic realm of technology, various container technologies offer efficient deployment and resource utilization in edge devices. However, limited research has explored how various container ...technologies perform in specific domains. In response, this paper addresses this gap by evaluating container technologies like RunC, LXC, Containerd, Docker, Podman, and Singularity in OpenCV-based computer vision applications on ARM-based edge devices. Results show comparable performance between containerized and non-containerized applications. Containerd excels in memory reading, with both Containerd and LXC efficient in wired image reception, while Singularity and Containerd lead in wireless image reception. Despite Docker's slower memory reading, its consistently faster processing time positions it as a competitive option. Overall, Docker demonstrates superior efficiency for computer vision applications on ARM-based edge devices. These insights contribute to bridge the existing gap in integrating containers into IoT and ARM-based edge computing scenarios.
Wireless multimedia sensor networks can provide a much clearer picture of the sensed area and thus significantly improve many applications. However, the increased amount of data will lead to issues ...with energy consumption and network lifetimes if we use the traditional network-based data collection where packets are forwarded hop by hop to the base. To mitigate these problems, researchers have been investigating mobile sinks traveling through the network and collecting the generated data. Using mobile sinks significantly increases the delivery rate while also reducing energy usage, but this increases the collection delay because of the physical speed of mobile sinks. In this paper, to reduce the physical collection delay while maintaining the other performance improvements (e.g., delivery rate, energy usage), we focus on how to use faster mobile sinks. Such mobile sinks are often motion-constrained and require smooth paths which can be followed with these constraints. Accordingly, we first developed a basic smooth path construction algorithm based on the TSP. We then extended it with path adjustments based on the required contact time at each node. Finally, we allowed for multiple mobile sinks which allowed us to further reduce the average delay per packet over the previous algorithms. Through extensive simulations, we found our algorithm allows for faster data collection over the current solutions, which can only accommodate flexible but slower mobile sinks, while maintaining the increased delivery rate and decreased energy usage.
Software Defined Networking (SDN) has emerged as a promising paradigm to make network management easier while supporting various applications requiring different guarantees in terms of performance, ...availability and correctness. In essence, SDN decouples the control and data planes such that a logically centralized SDN controller finds the paths based on given requirements, and then accordingly installs the necessary forwarding rules on the distributed SDN switches, which are simple forwarding devices. Due to various reasons such as congestions, failures or policy changes in the network, the SDN controller needs to find new paths and shift the traffic from the original paths to the new paths. However, during this transition, numerous time consuming steps need to be performed such as updating nodes on the new paths. If not carefully coordinated, these updating steps may cause loops, black-holes, and performance degradations. In this paper, we provide a detailed survey of the existing techniques that try to consistently update the new state of the data plane. We also discuss open issues and future challenges.
Software Defined Networking (SDN) has emerged as a promising paradigm with the key idea of separating control and data planes. However, this separation causes various scalability issues (e.g., ...increased response time, overhead on the controller and core SDN switches) in Wide Area Networks (WANs). While addressing such scalability issues, it is important to maintain the performance gains of the Reactive SDN and also to handle link failures. To jointly address scalability, performance, and link failure issues, we propose a new Source-Path Routing Model (SPRM) by combining the best of proactive and reactive approaches along with a simple packet programming mechanism. Specifically, SPRM pre-calculates and pre-installs multiple paths (proactive part) but then selects/assigns one of these paths for an incoming flow based on the available dynamic network state information (NSI) (reactive part). We also utilize SPRM to effectively cope with the link failures by using multiple paths as backup to each other. Our simulation results show that SPRM performs much better than the pure Reactive SDN in terms of path establishment delay and flow table size while (a) maintaining its performance gains in terms of link utilization and load balancing, and (b) quickly recovering from a link failure without causing extra memory overhead on the data plane.
We start with role-based trust management (RBTM) and address some of the challenges associated with using RBTM in mobile ad hoc networks (MANETs). We then enhance RBTM with reputation systems (RSs), ...and propose a new hybrid trust management system (HTMS). In HTMS, the privilege level of an entity is determined not only by its role in the system, but also by its reputation score, which in turn is based on its behavior. If a privileged node becomes compromised and conducts several malicious or risky transactions, then its privilege level is quickly reduced to limit its access to resources and minimize the damage it can inflict further. The system uses a global, network-wide perspective to thwart global attacks. Such fine-grained variations of access control and dynamically assigning privilege levels would be very difficult to accomplish manually. We evaluated HTMS by comparing an implementation of it against an ideal response. We show that HTMS performs very close to the ideal if we can accurately estimate the proportion of malicious nodes in the network. We suggest using sampling to estimate this proportion. However, even if this estimate is not accurate, the results are still much better than using RBTM by itself.
EDICS:
SYS-ARCH; SYS-PROT; FOR-DETE; SYS-INTR.
Graph indexing and querying mechanisms have been receiving significant attention due to their importance in analyzing the growing graph datasets in many domains. Although much work has been done in ...the context of simple graphs, they are not directly applicable to hypergraphs that represent more complex relationships in various applications. The key problem here is to search a given subhypergraph query in a larger hypergraph dataset. This search problem is known to be NP-hard as it is related to graph isomorphism. To solve this search problem in an efficient manner, we first create an index set by extracting the common subhypergraph structures from the given hypergraph dataset. Upon receiving a query, we use the same indexing techniques and create a query index set from the given subhypergraph. Utilizing both indices, we identify the possible locations of the query in the hypergraph dataset. We then start the subhypergraph search to verify whether the query really appears at each location by using an accelerated verification mechanism called layer-related-closure method. Through experiments on a real hypergraph dataset and random datasets, we demonstrate the efficiency and effectiveness of hypergraph indexing and our verification method.
Routing in Software Defined Networking (SDN) requires the SDN controllers to obtain the current dynamic network state information (NSI) as accurately as possible while reducing protocol overhead. To ...balance the trade-offs between the accuracy and protocol overhead, the SDN controllers usually implement a periodic NSI collection design where they obtain the dynamic link-state information from each switch at a predetermined rate. To improve the accuracy of NSI while minimizing the protocol overhead and providing better load balancing among multi-controllers, we propose a link-prioritized NSI collection design and compare it against the existing periodic NSI collection design. Using a dynamic routing algorithm on RYU SDN controller and the Mininet emulator, we compare our design against existing periodic NSI collection design. We observe that the proposed link-prioritized NSI collection design significantly reduces the protocol overhead without sacrificing the routing performance in terms of throughput or accepted bandwidth request. Moreover, our design results in better packet loss because it can quickly and accurately obtain NSI.