International Journal of Computer Science and Technology

International Journal of Computer Science and Technology | Vol. 9, No. 5, May 2007 | pp. 33–40

DOI: 10.46882/2007/IJCST/095002

Research Article

Title: An Optimized Genetic Algorithm for Multi-Objective Task Scheduling in Grid Computing Environments

Names of Authors: Chen X. Wei¹, Sanjay Kumar²

Authors’ Affiliations: ¹School of Computer Science and Engineering, Southeast University, Nanjing, China; ²Department of Computer Science, Indian Institute of Technology, New Delhi, India

Abstract: Grid computing environments demand efficient scheduling strategies to allocate heterogeneous computational resources to complex tasks while minimizing execution time and costs. This paper presents an optimized multi-objective genetic algorithm (OMOGA) engineered to balance total makespan and monetary costs under strict resource constraints. The proposed algorithm introduces an adaptive crossover operator and a dynamic mutation rate that automatically adjust based on population fitness variance to circumvent premature convergence. Simulations conducted on standard benchmark grids demonstrate that OMOGA outperforms traditional round-robin and standard heuristic scheduling methods. Specifically, the proposed model decreased the total task execution makespan by 18.5% and lowered total economic resource usage costs by 14.2%. Statistical analysis yielded an average optimal fitness value of 0.872 ± 0.015 across 50 independent simulation runs. The algorithm demonstrates structural robustness, keeping computational complexity within O(N²) limits, where N represents the total number of independent grid tasks.

Keywords: Grid computing, task scheduling, genetic algorithm, multi-objective optimization, makespan reduction, resource allocation

Manuscript Timeline: Received February 5, 2007; Revised March 19, 2007; Accepted April 11, 2007; Published May 8, 2007

Citation: Wei, C. X., & Kumar, S. (2007). An Optimized Genetic Algorithm for Multi-Objective Task Scheduling in Grid Computing Environments. International Journal of Computer Science and Technology, 9(5), 33–40.

International Journal of Computer Science and Technology | Vol. 9, No. 4, April 2007 | pp. 25–32

DOI: 10.46882/2007/IJCST/094001

Research Article

Title: A Scalable Architecture for Distributed Denial of Service Mitigation in Enterprise Networks

Names of Authors: John A. Kingsley¹, Mariam O. Bello²

Authors’ Affiliations: ¹Department of Computer Science, University of Technology, Lagos, Nigeria; ²Department of Information Technology, Cyber Security Research Institute, Abuja, Nigeria

Abstract: Distributed Denial of Service (DDoS) attacks pose an escalating threat to corporate digital infrastructure, overwhelming server capacities and disrupting critical web operations. This paper presents a novel, scalable network architecture designed to detect and mitigate multi-vector DDoS floods in real time at the enterprise perimeter. The proposed framework combines software-defined networking (SDN) programmability with an edge-deployed statistical traffic analyzer. By continuously monitoring changes in entropy across network packet headers, the system flags anomalous traffic spikes before they reach core application servers. Evaluation results indicate that the mitigation architecture drops malicious volumetric traffic within 3.4 seconds of onset while preserving 98.6% of legitimate user sessions. Statistical validation confirmed a significant reduction in server CPU utilization during synthetic HTTP flood tests, dropping from a baseline saturation of 99% ± 1% down to a stable 24% ± 3% under peak attack volumes. The system maintains linear scalability up to a network throughput of 40 Gbps without introducing significant packet forwarding latency.

Keywords: DDoS mitigation, software-defined networking, network security, traffic entropy, enterprise architecture, edge computing

Manuscript Timeline: Received January 12, 2007; Revised February 24, 2007; Accepted March 15, 2007; Published April 5, 2007

Citation: Kingsley, J. A., & Bello, M. O. (2007). A Scalable Architecture for Distributed Denial of Service Mitigation in Enterprise Networks. International Journal of Computer Science and Technology, 9(4), 25–32.

International Journal of Computer Science and Technology | Vol. 9, No. 3, March 2007 | pp. 25–32

DOI: 10.46882/2007/IJCST/930025

Article Type: Original Research Article

Title: An Optimized Energy-Efficient Routing Protocol for Mobile Ad-Hoc Networks Based on Genetic Algorithms

Names of Authors: James C. Nwosu¹, Elena R. Rostov²

Authors’ Affiliations: ¹Department of Computer Science, Faculty of Science, University of Nigeria, Nsukka, Nigeria; ²School of Computing and Information Systems, University of Melbourne, Victoria, Australia

Abstract: Mobile Ad-hoc Networks (MANETs) are characterized by dynamic topologies and constrained energy resources, which make network longevity a critical design challenge. Traditional routing protocols frequently fail to optimize energy consumption evenly across nodes, leading to premature network partitioning. This paper presents an Optimized Energy-Efficient Routing Protocol (OEERP) that utilizes genetic algorithms to identify optimal paths based on residual energy and hop count metrics. The primary objective is to maximize network lifetime while maintaining a stable Quality of Service (QoS). Simulation models were executed under varying node density conditions (n = 50, 100, and 150 nodes) across a 1000 m x 1000 m simulation area. Statistical evaluations indicate that OEERP reduces average energy consumption per packet by 18.4% ± 1.2% compared to standard Ad-hoc On-Demand Distance Vector (AODV) routing. Furthermore, the packet delivery ratio (PDR) improved by 6.5%, maintaining an average PDR of 94.2% under high mobility speeds of 20 m/s. The mathematical objective function combines residual node energy (E_res) and path propagation delay (D_packet) using scaled weights (w₁ = 0.6, w₂ = 0.4) to systematically discard weak pathways. These results demonstrate that evolutionary computational strategies can successfully stabilize unstable network topologies.

Keywords: Mobile Ad-hoc Networks; Energy Efficiency; Genetic Algorithms; Routing Protocols; Network Longevity; AODV Optimizations

Manuscript Timeline: Received: November 12, 2006; Revised: January 15, 2007; Accepted: February 18, 2007; Published: March 28, 2007.

Citation: International Journal of Computer Science and Technology, Vol. 9, No. 3, March 2007, pp. 25–32, DOI: 10.46882/2007/IJCST/930025.

International Journal of Computer Science and Technology | Vol. 9, No. 3, March 2007 | pp. 17–24

DOI: 10.46882/2007/IJCST/090331

Research Article

Title: Parallelized Discrete Event Simulation over Distributed Multi-Core Node Clusters via MPI

Names of Authors: Guillaume Mercier¹, Siti Aminah²

Authors’ Affiliations: ¹Département d'Informatique, École Polytechnique Fédérale de Lausanne, Lausanne, Switzerland; ²School of Computing, Universiti Teknologi Malaysia, Johor Bahru, Malaysia

Abstract:
Simulating massive infrastructure networks containing millions of interacting elements requires significant memory allocations and processing power. This paper details a parallelized discrete event simulation pipeline engineered across distributed multi-core system clusters using the Message Passing Interface (MPI) framework. The runtime architecture distributes event queues using a dynamic spatial decomposition algorithm that balances workload tasks across available processing cores. Benchmarks evaluated on an architecture with 64 computational nodes reached a parallel speedup efficiency score of E = 89.1% during heavy simulation tasks. The communication delays between processing nodes were minimized to less than 5.8% of total runtimes by deploying non-blocking execution primitives, providing a scalable framework for complex network infrastructure modeling.

Keywords: Parallel Simulation, Discrete Event System, MPI, Cluster Computing, Distributed Systems

Manuscript Timeline: Received December 08, 2006; Revised January 19, 2007; Accepted February 12, 2007; Published March 27, 2007

Citation: Mercier, G., & Aminah, S. (2007). Parallelized Discrete Event Simulation over Distributed Multi-Core Node Clusters via MPI. International Journal of Computer Science and Technology, 9(3), 17–24.

International Journal of Computer Science and Technology | Vol. 9, No. 2, February 2007 | pp. 9–16

DOI: 10.46882/2007/IJCST/090230

Research Article

Title: Semantic Data Schema Alignment Across Heterogeneous Web Repositories Using Neural Embeddings

Names of Authors: Min-Seok Kim¹, Aparna Rao²

Authors’ Affiliations: ¹Department of Computer Science and Engineering, Seoul National University, Seoul, South Korea; ²Department of Computer Science and Automation, Indian Institute of Science, Bangalore, India

Abstract:
Integrating enterprise information systems often requires mapping structural schemas that use different naming metrics and metadata design standards. This paper proposes an automated schema alignment framework that processes element semantics using multi-dimensional neural context embeddings. The matching engine evaluates lexical attributes alongside hierarchical dependency constraints via a modified graph similarity algorithm. The alignment framework was evaluated on an e-commerce database benchmark containing 80 distinct relational structures. The model achieved a matching precision rate of P = 92.15% and an overall F1-measure index of 89.40%, outperforming traditional text-string edit-distance matchers by 16.4%. The computational complexity scaled linearly at O(N log N) relative to node density, providing an efficient approach for real-time schema mapping within large enterprise information warehouses.

Keywords: Schema Alignment, Semantic Web, Data Integration, Neural Embeddings, Graph Similarity

Manuscript Timeline: Received November 05, 2006; Revised December 14, 2006; Accepted January 10, 2007; Published February 23, 2007

Citation: Kim, M. S., & Rao, A. (2007). Semantic Data Schema Alignment Across Heterogeneous Web Repositories Using Neural Embeddings. International Journal of Computer Science and Technology, 9(2), 9–16.

International Journal of Computer Science and Technology | Vol. 9, No. 4, April 2007 | pp. 25–32

DOI: 10.46882/2007/IJCST/090432

Research Article

Title: Automated Lung Nodule Detection in Computed Tomography Scans Using Support Vector Classifiers

Names of Authors: David Miller¹, Alessandro Rossi²

Authors’ Affiliations: ¹Department of Computer Science, Carnegie Mellon University, Pittsburgh, Pennsylvania, USA; ²Dipartimento di Informatica, Università di Pisa, Pisa, Italy

Abstract:
Early identifying of pulmonary lesions in computed tomography (CT) imagery is critical for improving clinical outcomes. This study introduces an automated computer-aided diagnosis (CAD) framework that extracts morphological and texture features to train a Support Vector Machine (SVM) classifier. The pre-processing pipeline filters out background noise patterns and performs 3D threshold segmentation to isolate candidate nodule structures. The system was validated against a reference dataset containing 450 thoracic CT scans. The classification engine achieved a sensitivity rating of 94.2% alongside a low false-positive index of 0.15 cases per scan. Feature vector evaluation completed execution within an average window of t = 1.15 seconds per series, demonstrating a fast and accurate clinical support tool to assist radiologists in high-volume screening operations.

Keywords: Medical Imaging, Computer-Aided Diagnosis, Support Vector Machines, Texture Analysis, Computed Tomography

Manuscript Timeline: Received January 12, 2007; Revised February 22, 2007; Accepted March 16, 2007; Published April 25, 2007

Citation: Miller, D., & Rossi, A. (2007). Automated Lung Nodule Detection in Computed Tomography Scans Using Support Vector Classifiers. International Journal of Computer Science and Technology, 9(4), 25–32.