Effective kernel mapping for OpenCL applications in heterogeneous platforms

dc.citation.epage88en_US
dc.citation.spage81en_US
dc.contributor.authorAlbayrak, Ömer Erdilen_US
dc.contributor.authorAktürk, İsmailen_US
dc.contributor.authorÖztürk, Özcanen_US
dc.coverage.spatialPittsburg, PA, USAen_US
dc.date.accessioned2016-02-08T12:10:13Z
dc.date.available2016-02-08T12:10:13Z
dc.date.issued2012-09en_US
dc.departmentDepartment of Computer Engineeringen_US
dc.descriptionConference name: International Conference on Parallel Programming
dc.descriptionDate of Conference: September 2012
dc.description.abstractMany core accelerators are being deployed in many systems to improve the processing capabilities. In such systems, application mapping need to be enhanced to maximize the utilization of the underlying architecture. Especially in GPUs mapping becomes critical for multi-kernel applications as kernels may exhibit different characteristics. While some of the kernels run faster on GPU, others may refer to stay in CPU due to the high data transfer overhead. Thus, heterogeneous execution may yield to improved performance compared to executing the application only on CPU or only on GPU. In this paper, we propose a novel profiling-based kernel mapping algorithm to assign each kernel of an application to the proper device to improve the overall performance of an application. We use profiling information of kernels on different devices and generate a map that identifies which kernel should run on where to improve the overall performance of an application. Initial experiments show that our approach can effectively map kernels on CPU and GPU, and outperforms to a CPU-only and GPU-only approach. © 2012 IEEE.en_US
dc.description.provenanceMade available in DSpace on 2016-02-08T12:10:13Z (GMT). No. of bitstreams: 1 bilkent-research-paper.pdf: 70227 bytes, checksum: 26e812c6f5156f83f0e77b261a471b5a (MD5) Previous issue date: 2012en
dc.identifier.doi10.1109/ICPPW.2012.14en_US
dc.identifier.issn1530-2016en_US
dc.identifier.urihttp://hdl.handle.net/11693/28069en_US
dc.language.isoEnglishen_US
dc.publisherInstitute of Electrical and Electronics Engineersen_US
dc.relation.isversionofhttp://dx.doi.org/10.1109/ICPPW.2012.14en_US
dc.source.titleInternational Conference on Parallel Programmingen_US
dc.subjectGPUen_US
dc.subjectHeterogeneousen_US
dc.subjectKernelen_US
dc.subjectMappingen_US
dc.subjectOpenCLen_US
dc.subjectApplication mappingen_US
dc.subjectGPUen_US
dc.subjectHeterogeneous platformsen_US
dc.subjectKernel mappingen_US
dc.subjectMany coreen_US
dc.subjectMulti-kernelen_US
dc.subjectProcessing capabilityen_US
dc.subjectConformal mappingen_US
dc.subjectData transferen_US
dc.subjectProgram processorsen_US
dc.subjectEmbedded systemsen_US
dc.titleEffective kernel mapping for OpenCL applications in heterogeneous platformsen_US
dc.typeConference Paperen_US

Files

Original bundle

Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
Effective kernel mapping for OpenCL applications in heterogeneous platforms.pdf
Size:
301.23 KB
Format:
Adobe Portable Document Format
Description:
Full printable version