Matches in SemOpenAlex for { <https://semopenalex.org/work/W2807725265> ?p ?o ?g. }
Showing items 1 to 82 of
82
with 100 items per page.
- W2807725265 abstract "Graphics Processing Unit (GPU) provides acceleration services to many applications, such as AI, games, media transcoding, etc. Virtualization on GPU is an enabling technology which facilitates the hardware resource sharing among multiple virtual machines (VMs). Sharing a GPU not only brings pros such as high utilization but also introduces cons such as resource contention and performance degradation. Although the existing GPU scheduling policies have been to some extent optimized, there are still some deficiencies, such as inefficient GPU sharing among multiple VMs, and high overhead within VM switching. As a result, the performance of GPU virtualization is limited by the current design, which lacks fine-grained scheduling supports. In this paper, we propose the Fine-grained schEduLing of vIrtualized gPu rEsources (FELIPE) to fully utilize and efficiently share a physical GPU among multiple VMs. To this end, we achieve the FELIPE optimization by introducing fine-grained scheduling mechanisms for virtualized GPU resources in three aspects: 1) We design a mixed time/event-based scheduling policy to reduce the idle time within VM switching. 2) We create a seamless VM assignment process, which enables VMs to switch seamlessly by stages. 3) We develop a hybrid per-ring/VM scheduling strategy, which schedules workloads to different GPU engines to run simultaneously. Then we implement the FELIPE with Intel Graphics Virtualization Technology for shared vGPU technology (GVT-g). Finally, the experimental evaluations show that the performance of the first two scheduling policies can respectively achieve up to 21.5% and 19.7% improvement, and the last one can improve the performance from 57.9% to 98.5% compared with the native design for two virtual machines." @default.
- W2807725265 created "2018-06-21" @default.
- W2807725265 creator A5023243107 @default.
- W2807725265 creator A5049487451 @default.
- W2807725265 creator A5080266922 @default.
- W2807725265 creator A5080928745 @default.
- W2807725265 date "2018-07-01" @default.
- W2807725265 modified "2023-09-27" @default.
- W2807725265 title "Efficient Sharing and Fine-Grained Scheduling of Virtualized GPU Resources" @default.
- W2807725265 cites W1966074204 @default.
- W2807725265 cites W2005574683 @default.
- W2807725265 cites W2015518316 @default.
- W2807725265 cites W2067231500 @default.
- W2807725265 cites W2076567851 @default.
- W2807725265 cites W2086782787 @default.
- W2807725265 cites W2100926301 @default.
- W2807725265 cites W2124365587 @default.
- W2807725265 cites W2133400870 @default.
- W2807725265 cites W2170532055 @default.
- W2807725265 cites W2313289305 @default.
- W2807725265 cites W2314312041 @default.
- W2807725265 cites W2322202675 @default.
- W2807725265 cites W2323693848 @default.
- W2807725265 cites W2342427079 @default.
- W2807725265 cites W2522765117 @default.
- W2807725265 cites W2608738900 @default.
- W2807725265 cites W2625231790 @default.
- W2807725265 cites W4235366964 @default.
- W2807725265 cites W4243173098 @default.
- W2807725265 doi "https://doi.org/10.1109/icdcs.2018.00077" @default.
- W2807725265 hasPublicationYear "2018" @default.
- W2807725265 type Work @default.
- W2807725265 sameAs 2807725265 @default.
- W2807725265 citedByCount "6" @default.
- W2807725265 countsByYear W28077252652019 @default.
- W2807725265 countsByYear W28077252652021 @default.
- W2807725265 countsByYear W28077252652022 @default.
- W2807725265 countsByYear W28077252652023 @default.
- W2807725265 crossrefType "proceedings-article" @default.
- W2807725265 hasAuthorship W2807725265A5023243107 @default.
- W2807725265 hasAuthorship W2807725265A5049487451 @default.
- W2807725265 hasAuthorship W2807725265A5080266922 @default.
- W2807725265 hasAuthorship W2807725265A5080928745 @default.
- W2807725265 hasConcept C111919701 @default.
- W2807725265 hasConcept C120314980 @default.
- W2807725265 hasConcept C162324750 @default.
- W2807725265 hasConcept C206729178 @default.
- W2807725265 hasConcept C21547014 @default.
- W2807725265 hasConcept C25344961 @default.
- W2807725265 hasConcept C2779851693 @default.
- W2807725265 hasConcept C41008148 @default.
- W2807725265 hasConcept C51332947 @default.
- W2807725265 hasConcept C513985346 @default.
- W2807725265 hasConcept C79974875 @default.
- W2807725265 hasConceptScore W2807725265C111919701 @default.
- W2807725265 hasConceptScore W2807725265C120314980 @default.
- W2807725265 hasConceptScore W2807725265C162324750 @default.
- W2807725265 hasConceptScore W2807725265C206729178 @default.
- W2807725265 hasConceptScore W2807725265C21547014 @default.
- W2807725265 hasConceptScore W2807725265C25344961 @default.
- W2807725265 hasConceptScore W2807725265C2779851693 @default.
- W2807725265 hasConceptScore W2807725265C41008148 @default.
- W2807725265 hasConceptScore W2807725265C51332947 @default.
- W2807725265 hasConceptScore W2807725265C513985346 @default.
- W2807725265 hasConceptScore W2807725265C79974875 @default.
- W2807725265 hasLocation W28077252651 @default.
- W2807725265 hasOpenAccess W2807725265 @default.
- W2807725265 hasPrimaryLocation W28077252651 @default.
- W2807725265 hasRelatedWork W2034396432 @default.
- W2807725265 hasRelatedWork W2075021084 @default.
- W2807725265 hasRelatedWork W2101313158 @default.
- W2807725265 hasRelatedWork W2121854243 @default.
- W2807725265 hasRelatedWork W2235294519 @default.
- W2807725265 hasRelatedWork W2270201953 @default.
- W2807725265 hasRelatedWork W2293452700 @default.
- W2807725265 hasRelatedWork W3014980064 @default.
- W2807725265 hasRelatedWork W3186546008 @default.
- W2807725265 hasRelatedWork W3124171372 @default.
- W2807725265 isParatext "false" @default.
- W2807725265 isRetracted "false" @default.
- W2807725265 magId "2807725265" @default.
- W2807725265 workType "article" @default.