Deep Neural Network Based Application Capacity Analysis in Finance System

The goal of system capacity analysis is to understand the current capacity usage and forecast future capacity impact based on various business scenarios. The successful capacity analysis is the key to identify the system bottleneck and plan for better resource allocation. However, IT systems for finance company are inherently large and complex with numerous interfaces with other systems. Thus, identifying and selecting a good model to describe the system interdependence from capacity perspective is important but challenging problem. In our paper, we first define the problem we want to solve. We discuss 2 approaches as baselines. Then we propose DNN based multiple linear regression, which is more efficient for complex finance systems. We collected 12 months real production volume data as our dataset. The experiment shows our proposed model can give a better performance compared with baseline approaches. Unlike other research papers, our proposal focuses to solve problem in real finance industry.