Meeting Its Potential: A Cost-Sensitive Approach for Resource Allocation in Virtual Machines
Throughout the recent years, ING has made a shift from physical servers to virtual machines cluster warehouses for its IT units and use cases. While this transition has contributed to ING’s development teams in providing agility in requirements and elasticity in resource allocations, the potential for cost reduction on infrastructure spending has not fully been realized. Many virtual machines have not been shifting their resource allocation actively with their utilization pattern, which resulted that a large portion of over 60M EUR spent yearly for ING-DBNL’s infrastructure goes to idle computing infrastructure. Dor will discuss the factors teams evaluate when faced with a cost reduction request and discuss his analysis and modelling that enables teams make better decisions on their infrastructure setup.
Dor has over a decade of experience developing big data products for security industries, financial markets and banking industries. He has earned his masters’ degree as a McDonnell leadership scholarship recipient, working in the machine learning research group in Washington University. His work on metric learning and cost-sensitive learning has earned him publications in NIPS, AISTATS and a monetary prize in Cha-Learn competitions. As a data scientist at ING domestic banking, he is involved with multiple projects modelling consumer and market behavior and optimizing virtual environments. He is currently completing an MBA on a Big Data Track from Amsterdam University.