Ajwani, DeepakDeepakAjwaniAli, ShoukatShoukatAliMorrison, John P.John P.Morrison2019-04-112019-04-112012 IEEE2012-08-16978-1-4673-0975-21530-2075http://hdl.handle.net/10197/9900The 2012 IEEE International Parallel and Distributed Symposium (IPDPS), 21-25 May 2012, Shanghai, ChinaOptical circuit switches have recently been proposed as a low-cost, low-power and high-bandwidth alternative to electronic switches for the design of high-performance compute clusters. An added advantage of these switches is that they allow for a reconfiguration of the network topology to suit the requirements of the application. To realize the full potential of a high-performance computing system with a reconfigurable interconnect, there is a need to design algorithms for computing a topology that will allow for a high-throughput load distribution, while simultaneously partitioning the computational task graph of the application for the computed topology. In this paper, we propose a new framework that exploits such reconfigurable interconnects to achieve these interdependent goals, i.e., to iteratively co-optimize the network topology configuration, application partitioning and network flow routing to maximize throughput for a given application. We also present a novel way of computing a high-throughput initial topology based on the structural properties of the application to seed our co-optimizing framework. We show the value of our approach on synthetic graphs that emulate the key characteristics of a class of stream computing applications that require high throughput. Our experiments show that the proposed technique is fast and computes high-quality partitions of such graphs for a broad range of hardware parameters that varies the bottleneck from computation to communication.en© 2012 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.Graph-partitioning algorithmsReconfigurable topologyOptical circuit switchCo-optimizationConference Publication83684710.1109/IPDPS.2012.802019-04-01https://creativecommons.org/licenses/by-nc-nd/3.0/ie/