Distributed Texture Memory in a Multi-GPU Environment

dc.contributor.authorMoerschell, Adamen_US
dc.contributor.authorOwens, John D.en_US
dc.date.accessioned2015-02-21T12:52:50Z
dc.date.available2015-02-21T12:52:50Z
dc.date.issued2008en_US
dc.description.abstractIn this work, we demonstrate a system that allows texture memory on multiple graphics processing unit (GPUs) to be virtualized in a manner that is both scalable and transparent to the programmer. Our system is built using a directory-based shared-memory abstraction to allow texture memory to be distributed while staying consistent. We use texture pages as our basic memory block and discuss the data structures, threading model, and consistency mechanisms necessary to implement a paging system in a multi-GPU environment. The system is demand-driven, and pages will only be loaded into the texture memory of a GPU that makes a request. The main contribution of this work is the identification of the mechanisms required to implement our abstraction, as well as the discussion of its limitations in order to make it more efficient.en_US
dc.description.number1en_US
dc.description.seriesinformationComputer Graphics Forumen_US
dc.description.volume27en_US
dc.identifier.doi10.1111/j.1467-8659.2007.01101.xen_US
dc.identifier.issn1467-8659en_US
dc.identifier.pages130-151en_US
dc.identifier.urihttps://doi.org/10.1111/j.1467-8659.2007.01101.xen_US
dc.publisherThe Eurographics Association and Blackwell Publishing Ltden_US
dc.titleDistributed Texture Memory in a Multi-GPU Environmenten_US
Files
Collections