Mean-Field Limits of Randomized Load Balancing Networks

Mon, Aug. 31

3-4PM

400 Cory

3-4PM

400 Cory

We establish mean-field limits for a class of randomized load balancing models in networks with many servers in the presence of general service distributions.
Key components of the approach include a convenient representation of the state process, and the derivation of a system of partial differential equations (PDEs) that approximate important performance measures of the system. We validate the PDE approximation against simulations, and also describe how the PDEs can be used to identify some non-intuitive properties of the system. This is joint work with Mohammadreza Aghajani and Xingjie (Helen) Li.

Bio: Kavita Ramanan is a professor at the Division of Applied Mathematics at Brown University. Prior to Brown University, she held positions as a professor in the Division of Mathematical Sciences at Carnegie Mellon University and as a Member of Technical Staff at Lucent Bell Laboratories. She was awarded the Erlang Prize of the INFORMS Applied Probability Society and a Medallion from the IMS (Institute for Mathematics and Statistics). She is also a Fellow of the IMS. Her research lies in the area of probability theory and stochastic processes, including stochastic analysis, large deviations, Gibbs measures, measure-valued processes and applications to stochastic networks.

Bio: Kavita Ramanan is a professor at the Division of Applied Mathematics at Brown University. Prior to Brown University, she held positions as a professor in the Division of Mathematical Sciences at Carnegie Mellon University and as a Member of Technical Staff at Lucent Bell Laboratories. She was awarded the Erlang Prize of the INFORMS Applied Probability Society and a Medallion from the IMS (Institute for Mathematics and Statistics). She is also a Fellow of the IMS. Her research lies in the area of probability theory and stochastic processes, including stochastic analysis, large deviations, Gibbs measures, measure-valued processes and applications to stochastic networks.

UC Berkeley Networking

Ashwin Pananjady and Orhan Ocal

Last Modification Date: Wednesday, February 10, 2016

Ashwin Pananjady and Orhan Ocal

Last Modification Date: Wednesday, February 10, 2016