Conference Papers

Permanent URI for this collectionhttps://idr.nitk.ac.in/handle/123456789/28506

Browse

Has files: NoSubject: search.filters.subject.Resource Allocation

Search Results

Now showing 1 - 2 of 2
  • No Thumbnail Available
    Item
    Fairness in CPU Scheduling: A Probabilistic Algorithm
    (Institute of Electrical and Electronics Engineers Inc., 2024) Prasanna, S.; Gulati, A.; Anagha, H.C.
    This paper introduces a novel CPU scheduling algorithm for uniprocessor systems that employs a probabilistic function to enhance fair resource allocation. Unlike traditional algorithms, our approach specifically tackles the challenge of equitable resource distribution by integrating a probabilistic methodology whilst also keeping the priority of each process in mind. We detail the implementation and evaluate its performance against established algorithms, assessing metrics such as average turnaround time, average waiting time and the gini index. All the related code, data used for testing and a working webpage to try out the algorithm first hand can be found at GitHub. © 2024 IEEE.
  • No Thumbnail Available
    Item
    Optimization of Resource and Energy in Distributed Systems Using Unified Genetic Algorithm
    (Institute of Electrical and Electronics Engineers Inc., 2025) Dhruthi, G.; Sinchana, N.M.; Annappa, B.; Kumar, N.M.R.
    Cloud and large distributed systems must ensure resource scheduling, energy management, and resource allocation. However, there exist complex and dynamic workloads, which may cause inefficient resource distribution, increased energy consumption, cost of operation and time delays which ultimately lead to reduced Quality of Experience (QoE). To address these issues the Unified Genetic Algorithm (UGA) is proposed, a proactive approach in optimization which helps achieve relatively better balance between CPU and memory usage across multiple nodes in a distributed system. UGA, was tested using the Materna workload trace and subjected to comparison with other existing load balancing algorithms such as Firefly, Coral Reef Optimization and Novel Family. It is found that UGA is superior with regard to efficiency in scheduling as it has revealed an improvement of 6.72% in average when compared to state of the art algorithms and proved to be beneficial in optimal resource allocation and improvement in system performance. © 2025 IEEE.