| TITLE |
Transforming Static Server Allocation into an Adaptive Compute for Enhanced Throughput and SLA Compliance |
| ABSTRACT |
The research suggests a system of managing compute resource hardware management called Priority-Based Optimized Compute Resource Management to improve batch processing efficiency in a computing context. It manages compute resources across a series of data centers and disaster recovery sites, combining an adaptive microservices layer and both AutoSys and Univa Grid Engine schedulers. It also has a method of task and job prioritization based on multi-factor algorithms assessing workloads on the cluster, criticality of applications and client implications. While AutoSys prioritizes the jobs, the microservices layer optimizes resources dynamically. Univa Grid Engine optimizes scheduling based on server conditions. The framework commits to data analytics and machine learning for monitoring and in-flight modification of it resource management being the biggest factor in manage workload when accomplish the outcomes of increase compute efficiency (35-40% improvement), I.T infrastructure utilization (25% improved), reduced SLA violations all without adding any additional hardware. As a methodology it takes the static enterprise batch system and turns it into a structured, dynamic self-optimization system to allow for better workload orchestration and ability to adapt in cloud environments. |
| AUTHOR |
Nagabhushanam Bheemisetty
Senior Architect, Capgemini Financial Services USA Inc., USA |
| PUBLICATION DATE |
2025-11-19 19:22:14 |
| VOLUME |
13 |
| ISSUE |
2 |
| DOI |
DOI: 10.15662/IJMSERH.2025.1302080 |
| PDF |
pdf/2025/4/80_Transforming Static Server Allocation into an Adaptive Compute for Enhanced Throughput and SLA Compliance.pdf |
| KEYWORDS |
|
