MUMBAI, India, Sept. 28 -- Intellectual Property India has published a patent application (202621093525 A) filed by Dr. Chandra Shekhar Gautam; Dr. Laxmi Narayan Soni; Dr. Poonam Bhartiya; Pragya Shrivastava; Rashmi Rani Gautam; Aarti Singh Parihar; and Arjita Singh on August 01, 2026, for System And Method For Optimizing Query Execution Performance In Hadoop Mapreduce Environments Using Genetic Algorithms.
Inventors include Dr. Chandra Shekhar Gautam; Dr. Laxmi Narayan Soni; Dr. Poonam Bhartiya; Pragya Shrivastava; Rashmi Rani Gautam; Aarti Singh Parihar; and Arjita Singh.
The application for the patent was published on September 25, 2026, under issue no. 39/2026.
Abstract: ABSTRACT System and Method for Optimizing Query Execution Performance in Hadoop MapReduce Environments Using Genetic Algorithms The present disclosure relates to a system and method for optimizing query execution performance in Hadoop MapReduce environments using Genetic Algorithms. The disclosed system intelligently analyzes incoming queries together with real-time Hadoop cluster information to generate multiple candidate execution plans. Hadoop execution parameters, including mapper and reducer allocation, memory configuration, scheduling policy, block size, and data partitioning strategy, are encoded as chromosomes and iteratively optimized through Genetic Algorithm operations comprising fitness evaluation, selection, crossover, and mutation. This optimization process enables the identification of an efficient execution configuration for distributed data processing. The proposed system employs a multi-objective fitness function that simultaneously evaluates execution time, processor utilization, memory consumption, disk input/output performance, network communication overhead, data locality, load balancing, and energy consumption. Based on the computed fitness values, the optimization engine automatically selects the most suitable execution plan and deploys the optimized configuration to the Hadoop MapReduce framework. The system continuously adapts to varying workloads and heterogeneous cluster conditions, thereby eliminating the need for manual parameter tuning and improving operational efficiency. The disclosed invention further maintains a historical execution repository for adaptive learning and continuous optimization of future query executions. By leveraging historical performance data along with real-time cluster monitoring, the system enhances scalability, resource utilization, throughput, and reliability of Hadoop clusters. The invention is applicable to cloud computing, big data analytics, financial services, healthcare, scientific research, Internet of Things (IoT), smart cities, and other data-intensive applications requiring efficient distributed query processing.
Disclaimer: Curated by HT Syndication.
