Show simple item record

Files in this item

Thumbnail

Item metadata

dc.contributor.authorVarghese, Blesson
dc.contributor.authorPrades, Javier
dc.contributor.authorReaño, Carlos
dc.contributor.authorSilla, Federico
dc.date.accessioned2015-10-05T14:40:08Z
dc.date.available2015-10-05T14:40:08Z
dc.date.issued2015-09-01
dc.identifier.citationVarghese , B , Prades , J , Reaño , C & Silla , F 2015 , Acceleration-as-a-Service : exploiting virtualised GPUs for a financial application . in 2015 IEEE 11th International Conference on e-Science (e-Science) (2015) . IEEE Computer Society , 11th IEEE International Conference on eScience , Munich , Germany , 31/08/15 .en
dc.identifier.citationconferenceen
dc.identifier.otherPURE: 221662230
dc.identifier.otherPURE UUID: 87604533-39b8-4445-ae1d-055c1a8f959a
dc.identifier.otherWOS: 000380433500006
dc.identifier.urihttp://hdl.handle.net/10023/7601
dc.descriptionThis research was supported by an NVIDIA award, the EPSRC EP/K015745/1 grant, and the Generalitat Valenciana PROMETEOII/2013/009 grant of the PROMETEO program phase II. The authors acknowledge the generous support of Mellanox Technologies.en
dc.description.abstractHow can GPU acceleration be obtained as a service in a cluster? This question has become increasingly significant due to the inefficiency of installing GPUs on all nodes of a cluster. The research reported in this paper is motivated to address the above question by employing rCUDA (remote CUDA), a framework that facilitates Acceleration-as-a-Service (AaaS), such that the nodes of a cluster can request the acceleration of a set of remote GPUs on demand. The rCUDA framework exploits virtualisation and ensures that multiple nodes can share the same GPU. In this paper we test the feasibility of the rCUDA framework on a real-world application employed in the financial risk industry that can benefit from AaaS in the production setting. The results confirm the feasibility of rCUDA and highlight that rCUDA achieves similar performance compared to CUDA, provides consistent results, and more importantly, allows for a single application to benefit from all the GPUs available in the cluster without loosing efficiency.
dc.format.extent10
dc.language.isoeng
dc.publisherIEEE Computer Society
dc.relation.ispartof2015 IEEE 11th International Conference on e-Science (e-Science) (2015)en
dc.rights© 2015. IEEE. This is the accepted manuscript of a conference paper originally submitted to the 11th IEEE International Conference on eScience available from http://www.computer.org/csdl/proceedings/escience/index.htmlen
dc.subjectrCUDAen
dc.subjectGPU computingen
dc.subjectVirtualisationen
dc.subjectAcceleration-as-a-serviceen
dc.subjectCUDAen
dc.subjectQA75 Electronic computers. Computer scienceen
dc.subjectNDASen
dc.subject.lccQA75en
dc.titleAcceleration-as-a-Service : exploiting virtualised GPUs for a financial applicationen
dc.typeConference itemen
dc.description.versionPostprinten
dc.contributor.institutionUniversity of St Andrews.School of Computer Scienceen
dc.identifier.urlhttp://www.computer.org/csdl/proceedings/escience/index.htmlen


This item appears in the following Collection(s)

Show simple item record