Modern mobile applications are increasingly relying on cloud data centers to obtain both compute and storage capabilities. To guarantee good performance for customers, cloud platforms usually provision resources dynamically in order to meet the demand of the fluctuating workload. Cloud providers also price their services dynamically to meet variable demand. Mobile workloads exhibit three distinct characteristics: geographic or spatial fluctuation, rapid changes characterized by a shorter time scale of fluctuation, and more frequent fluctuations, which make current resource provisioning approaches less effective.
The MOBILESCALE project proposes new research on resource management for mobile workload to address dual objectives of handling fluctuations and optimizing cost and performance by leveraging application-specific optimization approaches, acknowledging the cost anomalies of cloud transient servers, and unifying resources from multiple cloud providers. The proposed work will involve the design and implementation of a mobile-aware resource management system, a set of mobile application benchmarks, and experimental evaluation of the proposed mobile-aware provisioning algorithms using home-grown mobile apps developed in this project.
The project will open doors for research on making modern cloud platforms more intelligent with tailored application-specific optimizations. The MOBILESCALE project can be used and integrated with current general-purpose resource management systems to reduce manual intervention from operators and to mask operational complexity. The broader impacts of the project will include graduate and undergraduate courses that incorporate research results, outreach to expose undergraduates, high school and K-12 students to research in computer systems, and the release of MOBILESCALE source code to the research community through the project website.
This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
