International Journal of Cloud Computing (8 papers in press)

Regular Issues

• Improved task scheduling strategy for balancing resource utilization and service quality in mobile edge computing environment  
  by Michael Pendo John Mahenge  
  Abstract: The rapid growth of resource-hungry and time-critical applications reflects the rise of resources needed for communication, processing, and energy consumption. Mobile edge computing (MEC) that offers cloud-computing services proximate to users at the edge of mobile network is considered to be the key technology to facilitate task scheduling closer to data-sources. The objective of this paper is to propose an improved task scheduling strategy that selects the best MEC server to process each task which reduces the system energy consumption and delay. Therefore, we proposed an improved task scheduling strategy on the basis of non-dominated sorting genetic algorithm II (NSGA-II). To improve the performance of NSGA-II, we proposed a hierarchical search policy (NSGA-H) that eliminates the number of redundant comparisons and thus, enhances time complexity. The simulation results illustrate that the proposed strategy improves average service delay, service time, energy consumption, and system utility compared to baseline approaches.
  Keywords: mobile edge computing; MEC; task scheduling; quality of experience; QoE; resource-intensive tasks; non-dominated sorting genetic algorithm II; NSGA II.
  DOI: 10.1504/IJCC.2024.10058683
   
  
• Cloud based scalable resiliency pattern using PRISM  
  by Punithavathy Ellappan, Priya N 
  Abstract: Applications in distributed systems are enhanced due to microservice architecture. It enriches the cloud's unique features like availability and scalability. The distributed nature has a broad set of failure points; thereupon resilience is the predominant factor for surviving these failures. Resilient feature of a microservice-based application is substantially offered by circuit breaker pattern, which scans the failure rate and safeguards from cascading failures. This paper analyses the behaviour pattern of a microservice based application under transient failure. As a result, the execution time during failure of an application is 23% faster when working with internal circuit breakers. Model-based verification techniques such as CTMC were performed to analyse the steady state probability of completed requests between the working cases of internal circuit breakers and proxy circuit breakers. The generated probability values of the internal circuit breaker, assure the 99% availability of the service even at times of failure.
  Keywords: circuit breaker; resiliency; microservices; cascading failures; continuous-time Markov chain; CTMC; PRISM.
  DOI: 10.1504/IJCC.2024.10058869
   
  
• An Intelligent Blockchain based Cryptographic Data Security (IBCDS) Model for an Efficient Data Sharing in Cloud  
  by Ponnada Naga Ramya, Ravi Prakash Reddy, Supreethi KP 
  Abstract: Secure data sharing is the most challenging and essential problem to be addressed in cloud systems. In traditional works, various blockchain and cryptographic approaches are deployed for enabling secured data storage and retrieval in cloud platform. However, the conventional frameworks require third-party entities for user authentication, data verification, and identity management. In our proposed work, an intelligent blockchain-based cryptographic data security (IBCDS) scheme is developed, where no third-party auditor is required for authentication and validation for secured data sharing over cloud systems with minimal overhead and computational complexity. In IBCDS, blockchain acts like user management system that stores information of each cloud user in the form of transactions by assuring reliability and integrity of each transaction along with tamper-proof. During analysis, IBCDS mechanism is evaluated and compared by using different parameters where security performance is improved to 99%, time complexity reduced to 98%, and overall throughput is maximised to 99%.
  Keywords: cloud systems; data security; blockchain; cryptography; transactions; encryption; decryption.
  DOI: 10.1504/IJCC.2024.10059968
   
  
• Load Balancing Using Improved Weighted Round Robin Algorithm in Cloud Computing Environment  
  by Sree Priya S, T. Rajendran 
  Abstract: Load balancing strategies maximise resource use, system efficiency, reliability, high access, network traffic management, and response to changing circumstances. This abstract provides the improved weighted round robin (IWRR) algorithm for cloud computing load balancing. Cloud infrastructure relies on load balancing to optimise resource use and server performance. IWRR dynamically adjusts server weights depending on real-time performance parameters like CPU utilisation and request latency, improving weighted round robin. Load-balancing solutions like the IWRR algorithm may improve cloud infrastructure scalability, dependability, and performance. Load balancing methods like round robin, IP hash and weighted round robin help distribute internet traffic across servers. For requests from the same domain name, IP hash is used, and balanced round robin is used when server capacity allows. These methods can be assessed by reaction time and capacity. Weighted round robin (WRR) dynamically assigns requests by server capabilities to reduce response time and increase throughput. Automatically distributing more requests to capable servers improves system speed. These methods eliminate resource waste, enable scalability based on consumer demand, reduce disruptions, and improve client experience by uniformly dispersing jobs over multiple resources. Load balancing lets online service providers use their physical capabilities to create stable, adaptable, and excellent solutions.
  Keywords: load balancing; throughput; response time; round robin; IP hash; weighted round robin; WRR; resource allocation; improved intelligent infrastructures.
  DOI: 10.1504/IJCC.2024.10062149
   
  
• Virtual Machine Workload Prediction using Deep Learning  
  by Abhilash C. S, Chaithra Usha, Veena Garag, Priyanka H 
  Abstract: This paper presents a novel approach to optimise resource allocation in virtualised systems, aiming to maximise performance and minimise operational expenses. Leveraging deep learning models, specifically long-short-term memory (LSTM) and bidirectional gated recurrent unit (bi-GRU), the method focuses on forecasting CPU load patterns in virtual machines (VMs). Accurate predictions are crucial for proactive resource management in dynamic cloud-based infrastructures. LSTM and bi-GRU excel in handling time series forecasting due to their ability to detect temporal connections in sequential data. Using pre-processed historical CPU load data, the models undergo training with hyperparameter adjustments to enhance performance. Experimental results demonstrate that the proposed models outperform others, achieving lower average root mean square error (RMSE) values (0.05636) and mean absolute error (MAE) values (0.03721). Comparative analysis with LSTM, GRU, bi-LSTM, bi-GRU, LSTM-GRU, and bi-LSTM-GRU confirms the high predictive capabilities of LSTM and bi-GRU, with the bidirectional architecture of bi-GRU enhancing accuracy by capturing connections between previous and upcoming time steps.
  Keywords: virtual machines; VMs; long-short-term memory; LSTM; bi-GRU; CPU load prediction; cloud computing.
  DOI: 10.1504/IJCC.2024.10062593
   
  
• A Survey on Blockchain Architecture and Consensus Mechanism: Design Vulnerability and Security Analysis  
  by Shshikant Sharma, Dharmender Singh Kushwaha 
  Abstract: Today for an organisation, data security is the most crucial topic. An organisation needs to protect its information against cyberattacks. Cryptography, DLT, and blockchain technology provide higher security for data storage and prevent any cyberattack. The most prominent reasons for using this technology are its specific properties, such as decentralisation, transparency, autonomy (without human interaction), and robustness. This paper discusses the performance and limitations of the existing blockchain architecture, consensus mechanism, and the security aspect of the consensus mechanism in an organised way. This survey presents systematic reviews of blockchain architectures, consensus mechanisms, the performance analysis of the current blockchain consensus mechanisms, and the vulnerabilities and types of attacks. The aim is to provide a comprehensive state-of-the-art platform where a beginner can swiftly move on to research aspects.
  Keywords: blockchain; distributed ledger technology; decentralisation; consensus mechanism; byzantine fault-tolerance; security.
  DOI: 10.1504/IJCC.2024.10062772
   
  
• Resource Scheduling in Cloud Environment using Particle Swarm Search algorithm  
  by Malay Kumar Majhi, Manas Ranjan Kabat, Satya Prakash Sahoo 
  Abstract: Cloud computing has gained significant popularity as a platform for processing large-scale data analytics, offering benefits such as high availability, robustness, and cost-effectiveness. However, job scheduling in cloud systems presents a major challenge, as it directly impacts execution time and operational costs. To address these issues, this paper presents a novel multi-adaptive convergent particle swarm optimisation (MAC-PSO) algorithm designed to decrease the failure rate, minimise makespan values, and enhance resource utilisation. The round Robin scheduling method aids in task execution by determining the appropriate time-space allocation. The proposed algorithm's performance is compared to that of the TLBO algorithm, demonstrating that MAC-PSO outperforms both TLBO and the original PSO. Moreover, a comprehensive analysis is proposed to evaluate the performance metrics within the MAC-PSO algorithm. Notably, MAC-PSO effectively increases the ratio of solutions that dominate previous algorithmic approaches and identifies a greater number of solutions that cater to user preferences.
  Keywords: task scheduling; particle swarm optimisation; PSO; round Robin scheduling; cloud computing.
  DOI: 10.1504/IJCC.2024.10064262
   
  
• AltWOA: Enhancing Query Performance with Clustering-Based Optimisation  
  by Mursubai Sandhya Rani, N.Raghavendra Sai 
  Abstract: Big data (BD) is gaining a lot of attention in the information field due to the data growth in the preceding ten years. A fundamental purpose of philosophical "query optimization (QO)" approaches in a BD environment is data retrieving. To offer beneficial and practical choices for BD query optimisation, numerous technologies that focus on the cloud have been developed. Existing significant data query optimisation approaches often struggle to efficiently process complex queries on massive datasets, leading to performance bottlenecks and resource wastage. Despite significant advancements in big data query optimisation, there remains a need for innovative techniques that can seamlessly handle diverse workloads and data distributions while optimizing resource utilisation and query performance. To solve query optimization issues, this paper suggests an Altruistic Whale Optimization Algorithm. In the following stage, the AltWOA optimizer increases the total query processing effectiveness while ignoring the energy-efficient query techniques. The metrics classification and computation time are tested for various data sizes, instances, and dataset records.
  Keywords: Big data (BD); query optimization (QO); Altruistic Whale Optimization Algorithm (AltWOA); fast Markov clustering algorithm.
  DOI: 10.1504/IJCC.2024.10064274