Leverage on Artificial Intelligence in Capacity Management: Predict IT assets usage based on Business events
The Societe Generale Bank possesses a network of trading applications which generate the hardware consumption data (CPU, memory and network communication). The main usage of this trading system is to receive/ send orders from the market and/ or the clients or vice versa, and based on these order information, business decisions are made accordingly.
The projects main research focus is to use Artificial Intelligence technologies to help better manage hardware capacity and data flow in the network.
To achieve this, we will be using different methodologies such as gradient boosting (and others) to investigate the relationship between the components of this network and make predictions on hardware consumption and how the whole network reacts to data flow and market fluctuation.
Ideally, this research will result in a clearer understanding of the network and help maintain the infrastructure and avoid system crush.