RECENT TRENDS IN PARALLEL COMPUTING https://stmcomputers.stmjournals.com/index.php/RTPC <p align="center"><strong>ISSN: 2393-8749</strong></p> <p align="center"><strong>Scientific Journal Impact Factor (SJIF):</strong> 6.112</p> <p><strong> </strong><strong> </strong><strong>Journal DOI no</strong>.: 10.37591/RTPC</p> STM JOURNALS ( Consortium eLearning Network Pvt Ltd) en-US RECENT TRENDS IN PARALLEL COMPUTING 2393-8749 Context-aware Multipath Load Balancing Routing for RPL in IoT https://stmcomputers.stmjournals.com/index.php/RTPC/article/view/629 <p><em>The Internet of Things (IoT) is revolutionizing the world with unprecedented levels of insights and intelligence, connecting the physical world to the digital age. In order to provide dependable and efficient routing, IPv6 Routing Protocol for Low Power and Lossy Networks (RPL) has been adopted as the foundation network for the Internet of Things (IoT). Although many IoT routing criteria have been met thanks to RPL's distinctive features, load balancing and congestion control are still the outliers. A context-aware Multipath Load Balancing Routing Protocol (CAMLB-RPL) is proposed that balances the traffic load across multiple paths, minimizing congestion, and increasing network reliability and lifetime. CAMLB-RPL leverages the Context-aware Load Balancing Routing metric (CALBR) to evaluate the queue occupancy and expected lifetime for the chain of parent nodes in the network. A context-aware Load Balancing Objective Function (CALB-OF) is also presented to identify the best-optimized load-balanced route during the parent selection process considering the CALBR and the Child count (CC) metric. A multipath routing strategy is employed to mitigate the congestion due to burst traffic in the network. The performance of CAMLB-RPL was compared to the typical RPL objective functions OF0 and MRHOF using the Cooja simulator of Contiki OS 3.0. CAMLB-RPL showed outstanding results, with power consumption decreasing by 39%, end-to-end delay decreasing by 36%, queue loss ratio declining by 68%, packet receiving rate increasing by 12.5%, and network lifetime increasing by 69%, on average.</em></p> Kala Venugopal T.G. Basavaraju Copyright (c) 2023 RECENT TRENDS IN PARALLEL COMPUTING 2023-08-17 2023-08-17 10 2 1 17 A Comprehensive Examination of Cloudflare's IP-based Distributed Denial of Service Mitigation https://stmcomputers.stmjournals.com/index.php/RTPC/article/view/665 <p class="Abstract" style="margin: 0in; line-height: normal;"><em><span style="font-size: 11.0pt;">This study dives deep into the world of DDoS (Distributed Denial of Service) attack prevention, with a keen focus on Cloudflare. As a powerhouse in the cybersecurity world, Cloudflare uses smart tactics, like acting as a reverse proxy, to keep online services safe from DDoS attacks. What does Cloudflare do exactly? It acts as a bouncer, checking all incoming server traffic to weed out harmful requests. This unique position lets Cloudflare effectively spot and deal with DDoS attacks. How does it do this? By looking at several factors, including the reputation of an IP, how often requests are coming in, and the information in HTTP headers. By doing this, Cloudflare can distinguish between legitimate user traffic and nasty attack traffic. What occurs when a DDoS attack is initiated? Cloudflare's robust setup jumps into action, soaking up the attack traffic and keeping it away from the target server. This means that regular users can keep accessing the server, keeping the service running without a hitch. One of the coolest things about Cloudflare's system is that it can scale up to deal with huge DDoS attacks that would otherwise crash a server's capacity. In this study, we will look at how Cloudflare's defense mechanisms against DDoS attacks work, particularly its system for managing IP reputation, how it limits the rate of requests, and how it inspects HTTP headers. We will also have a chat about the pros and cons of using Cloudflare for DDoS protection. The aim of all this? To provide solid insights for organizations looking to beef up their protection against DDoS attacks.</span></em></p> Muhammad Nadeem Syeda Wajiha Zahra Muhammad Nouman Abbasi Ali Arshad Saman Riaz Copyright (c) 2023 RECENT TRENDS IN PARALLEL COMPUTING 2023-10-21 2023-10-21 10 2 45 54 Addressing Security Challenges in Cloud Computing for Healthcare Systems: A Comprehensive Survey https://stmcomputers.stmjournals.com/index.php/RTPC/article/view/633 <p><em>Mobile cloud computing (MCC) emerges as an innovative technology revolutionizing mobile web services by combining the domains of mobile computing, cloud computing, and wireless networks. In the healthcare sector, MCC plays a pivotal role in transforming medical services, offering novel facilities and services to patients and caregivers alike. The adoption of MCC by healthcare professionals (HCPs) has led to a surge in medical software applications designed for these platforms, consequently enhancing medical services while reducing costs and fostering competition among healthcare providers. The integration of MCC has proven beneficial for HCPs, aiding them in making informed decisions and providing improved patient care. Utilizing cloud computing services, MCC makes it easier to address the continuously expanding demands of the healthcare sector. However, this advancement has not been without challenges. Constraints related to network bandwidth and mobile device capacity have introduced concerns regarding energy consumption and latency delays. Numerous factors influence the performance of MCC systems, necessitating a comprehensive investigation to optimize their efficiency. Furthermore, addressing security and privacy issues associated with MCC in healthcare is of paramount importance. Despite its significant role in cloud computing, the domain lacks an in-depth study of these concerns. This study aims to bridge the gap by conducting a systematic analysis of 43 relevant papers published between 2010 and 2022 in reputable national and international journals, including Springer and IEEE. Through this study, we endeavor to explore the challenges faced by mobile cloud computing in healthcare and propose strategies to enhance healthcare services in the cloud environment.</em></p> Anjali Pawar Sunita Mahajan Ikvinderpal Singh Copyright (c) 2023 RECENT TRENDS IN PARALLEL COMPUTING 2023-08-17 2023-08-17 10 2 18 27 An Application of Recommendation Systems to Cluster-based Pattern Recognition https://stmcomputers.stmjournals.com/index.php/RTPC/article/view/668 <p><em>Information overload is a common problem faced by internet users; so, the recommendation system is </em><em>becoming increasingly important to this service. Collective filtering is used by recommendation systems to offer recommendations to individuals or for products based on their shared interests. Data or people should be grouped into clusters so that they have more in common with those in the same cluster and stand out from those in other clusters. Parameters like average, correlation, mutual information, etc. are used in pattern discovery and matching. To extract and organise patterns into categories, PR may be used as a classification approach. PR is a way of categorising. Public relations (PR) is a hybrid method. In this survey, we investigated data mining, clustering, recommendation systems, and pattern recognition. We defined numerous clustering and pattern recognition algorithms and discussed their applications.</em></p> Ankush Singh Bhavana Pillai Copyright (c) 2023 RECENT TRENDS IN PARALLEL COMPUTING 2023-10-21 2023-10-21 10 2 28 36 Cyberattack Detection System in Private Cloud https://stmcomputers.stmjournals.com/index.php/RTPC/article/view/614 <p><em>As cloud computing adoption in colleges continues to rise, the security of private cloud systems has become a paramount concern. Data breaches resulting from cyberattacks can inflict severe damage to a university's revenue and reputation. This research proposes a novel machine learning-based cyber threat detection system tailored to the university's private cloud environment. The system's main objective is to continuously monitor the cloud infrastructure and employ advanced machine learning algorithms to analyze network traffic, and identifying and preventing unusual activities that may indicate potential cyberattacks. By leveraging the potential of machine learning, this innovative system aims to enhance the university's cybersecurity protections. It considers the dynamic and evolving nature of cyber threats, enabling real-time detection and proactive measures against malicious activities. The integration of cutting-edge machine learning models and feature extraction techniques empowers the system to identify patterns of anomalous behavior, even in the face of sophisticated attacks. Essential elements of the suggested system encompass the analysis of network traffic, the identification of anomalies, and the incorporation of threat intelligence. Through the analysis of network packets and access logs, the system can effectively detect signs of unauthorized access, data exhilaration, and other cyber threats. Additionally, threat intelligence feeds provide the system with up-to-date information on emerging threats, enabling quick responses to potential risks. Moreover, the system's implementation adheres to privacy and data protection regulations, ensuring secure handling of sensitive information within the private cloud environment. Regular updates and adaptive learning capabilities enable the system to evolve with changing cyber threats, ensuring continued robustness in the face of new challenges. In conclusion, the proposed machine learning-based cyberattack detection system presents a powerful solution to safeguarding the university's private cloud infrastructure. By promptly detecting and mitigating potential cyber threats, the system acts as a proactive defense mechanism, safeguarding valuable data and preserving the university's reputation in the ever-evolving landscape of cyber security.</em></p> Chethan M.S. Girish L. Manjula T. Priya R. Acharya Nikki Kishore Chaitra M.S. Copyright (c) 2023 RECENT TRENDS IN PARALLEL COMPUTING 2023-10-09 2023-10-09 10 2 37 44