Abstract: IT workload distributions are affecting data centers un-suitable airflow and temperature within the distribution. Traditional Computational Fluid Dynamics (CFD) is time-consuming and expensive, computationally and financially while the Reduced Order Models (ROM’s) are fast as comparing towards. In AKS Extensible Catalog Framework (IAECF) contracting to ROM can do they without a user tolerance, and can make low errors using IAECF demonstrate for optimal workload distribution framework for processing in data centers. The optimization workload determination to lead approximately 8.2% (a year before comparing 2.4% growth) compared to energy savings with a particular uniform distribution with working on the basis of FFD improvement effective by the speed and multiplied-iteration for optimizing on data center eventually.The entire optimization process, including IAECF training in 3.2 hours in the design phase. This is an alternative methodology of CFD (based on Fast Fluid Dynamics) and preliminary study IAECF-based on optimizing algorithm placement platforms that are parallelizable and simple to program.

Key Words: Workload Optimization, Data Center, Online Learning, ROM Modules, Extensible Catalog Framework.

| DOI: 10.17148/IARJSET.2021.86125