Matches in SemOpenAlex for { <https://semopenalex.org/work/W4315705726> ?p ?o ?g. }
Showing items 1 to 69 of
69
with 100 items per page.
- W4315705726 abstract "With the increasing diversity of heterogeneous architecture in the HPC industry, porting a legacy application to run on different architectures is a tough challenge. In this paper, we present OpenMP Advisor, a first of its kind compiler tool that enables code offloading to a GPU with OpenMP using Machine Learning. Although the tool is currently limited to GPUs, it can be extended to support other OpenMP-capable devices. The tool has two modes: Training mode and Prediction mode. The training mode must be executed on the target hardware. It takes benchmark codes as input, generates and executes every variant of the code that could possibly run on the target device, and then collects data from all of the executed codes to train an ML-based cost model for use in prediction mode. However, in prediction mode the tool does not need any interaction with the target device. It accepts a C code as input and returns the best code variant that can be used to offload the code to the specified device. The tool can determine the kernels that are best suited for offloading by predicting their runtime using a machine learning-based cost model. The main objective behind this tool is to maintain the portability aspect of OpenMP. Using our Advisor, we were able to generate code of multiple applications for seven different architectures, and correctly predict the top ten best variants for each application on every architecture. Preliminary findings indicate that this tool can assist compiler developers and HPC application researchers in porting their legacy HPC codes to the upcoming heterogeneous computing environment." @default.
- W4315705726 created "2023-01-12" @default.
- W4315705726 creator A5011976150 @default.
- W4315705726 creator A5029915466 @default.
- W4315705726 creator A5050432598 @default.
- W4315705726 creator A5053733660 @default.
- W4315705726 date "2023-01-09" @default.
- W4315705726 modified "2023-10-16" @default.
- W4315705726 title "OpenMP Advisor" @default.
- W4315705726 doi "https://doi.org/10.48550/arxiv.2301.03636" @default.
- W4315705726 hasPublicationYear "2023" @default.
- W4315705726 type Work @default.
- W4315705726 citedByCount "0" @default.
- W4315705726 crossrefType "posted-content" @default.
- W4315705726 hasAuthorship W4315705726A5011976150 @default.
- W4315705726 hasAuthorship W4315705726A5029915466 @default.
- W4315705726 hasAuthorship W4315705726A5050432598 @default.
- W4315705726 hasAuthorship W4315705726A5053733660 @default.
- W4315705726 hasBestOaLocation W43157057261 @default.
- W4315705726 hasConcept C106251023 @default.
- W4315705726 hasConcept C111919701 @default.
- W4315705726 hasConcept C118524514 @default.
- W4315705726 hasConcept C13280743 @default.
- W4315705726 hasConcept C149635348 @default.
- W4315705726 hasConcept C169590947 @default.
- W4315705726 hasConcept C173608175 @default.
- W4315705726 hasConcept C177264268 @default.
- W4315705726 hasConcept C185798385 @default.
- W4315705726 hasConcept C199360897 @default.
- W4315705726 hasConcept C205649164 @default.
- W4315705726 hasConcept C2776760102 @default.
- W4315705726 hasConcept C2777904410 @default.
- W4315705726 hasConcept C41008148 @default.
- W4315705726 hasConcept C48677424 @default.
- W4315705726 hasConcept C63000827 @default.
- W4315705726 hasConcept C85687889 @default.
- W4315705726 hasConceptScore W4315705726C106251023 @default.
- W4315705726 hasConceptScore W4315705726C111919701 @default.
- W4315705726 hasConceptScore W4315705726C118524514 @default.
- W4315705726 hasConceptScore W4315705726C13280743 @default.
- W4315705726 hasConceptScore W4315705726C149635348 @default.
- W4315705726 hasConceptScore W4315705726C169590947 @default.
- W4315705726 hasConceptScore W4315705726C173608175 @default.
- W4315705726 hasConceptScore W4315705726C177264268 @default.
- W4315705726 hasConceptScore W4315705726C185798385 @default.
- W4315705726 hasConceptScore W4315705726C199360897 @default.
- W4315705726 hasConceptScore W4315705726C205649164 @default.
- W4315705726 hasConceptScore W4315705726C2776760102 @default.
- W4315705726 hasConceptScore W4315705726C2777904410 @default.
- W4315705726 hasConceptScore W4315705726C41008148 @default.
- W4315705726 hasConceptScore W4315705726C48677424 @default.
- W4315705726 hasConceptScore W4315705726C63000827 @default.
- W4315705726 hasConceptScore W4315705726C85687889 @default.
- W4315705726 hasLocation W43157057261 @default.
- W4315705726 hasOpenAccess W4315705726 @default.
- W4315705726 hasPrimaryLocation W43157057261 @default.
- W4315705726 hasRelatedWork W1499552465 @default.
- W4315705726 hasRelatedWork W2017260672 @default.
- W4315705726 hasRelatedWork W2081245617 @default.
- W4315705726 hasRelatedWork W2094976234 @default.
- W4315705726 hasRelatedWork W2365187805 @default.
- W4315705726 hasRelatedWork W2375332572 @default.
- W4315705726 hasRelatedWork W2382553934 @default.
- W4315705726 hasRelatedWork W2387090736 @default.
- W4315705726 hasRelatedWork W306278948 @default.
- W4315705726 hasRelatedWork W3103903655 @default.
- W4315705726 isParatext "false" @default.
- W4315705726 isRetracted "false" @default.
- W4315705726 workType "article" @default.