Sanjukta Das Smith discusses her research.
Published April 25, 2018
Researchers in the School of Management have developed a new algorithm that cloud computing service providers can use to establish pricing and allocate resources.
Forthcoming in Information Systems Research, the study provides practical formulas that companies like Amazon, Google or VMWare can use to determine the resources necessary to provide a certain level of service to their customers — and what to charge for it in their service-level agreements (SLA). In addition, any customer who uses cloud computing services can use the algorithm to help negotiate on pricing.
“As computing resources continue to shift to the cloud, customers are demanding near-constant access to their files and software, posing a significant challenge to service providers,” says study co-author Sanjukta Das Smith, associate professor of management science and systems. “A company can feed information about their data center into our algorithm, and out will come policy definitions and specific prescriptions on how to allocate resources and set pricing, all tailor-made to their environment.”
The researchers conducted extensive computational studies using real-world server log data to validate and supplement their analytical work.
“Better understanding of costs is crucial for effective resource provisioning,” says Smith. “Through our research, providers can fine-tune resources for each client, rather than the current strategy of relying on ‘guesstimates.’”
Smith collaborated on the study with Chunming Qiao, SUNY Distinguished Professor and chair of the Department of Computer Science and Engineering; Ram Ramesh, professor of management science and systems in the School of Management; and Shuai Yuan, PhD candidate in the School of Management. The study was funded by a grant from the National Science Foundation.