2026-07-23T15:20:47Z https://drops.dagstuhl.de/oai

oai:drops-oai.dagstuhl.de:10732 2024-03-06T11:05:05Z ddc:004 open_access

API Comparison of CPU-To-GPU Command Offloading Latency on Embedded Platforms (Artifact) Cavicchioli, Roberto Capodieci, Nicola Solieri, Marco Bertogna, Marko GPU Applications Heterogeneus systems High-performance heterogeneous embedded platforms allow offloading of parallel workloads to an integrated accelerator, such as General Purpose-Graphic Processing Units (GP-GPUs). A time-predictable characterization of task submission is a must in real-time applications. We provide a profiler of the time spent by the CPU for submitting stereotypical GP-GPU workload shaped as a Deep Neural Network of parameterized complexity. The submission is performed using the latest API available: NVIDIA CUDA, including its various techniques, and Vulkan. Complete automation for the test on Jetson Xavier is also provided by scripts that install software dependencies, run the experiments, and collect results in a PDF report. Schloss Dagstuhl – Leibniz-Zentrum für Informatik Roberto Cavicchioli and Nicola Capodieci and Marco Solieri and Marko Bertogna 2019 Is Part Of DARTS, Volume 5, Issue 1, Special Issue of the 31st Euromicro Conference on Real-Time Systems (ECRTS 2019) Article Text doc-type:ResearchArticle publishedVersion application/pdf doi:10.4230/DARTS.5.1.4 urn:nbn:de:0030-drops-107322 https://drops.dagstuhl.de/entities/document/10.4230/DARTS.5.1.4 eng https://creativecommons.org/licenses/by/3.0/de/legalcode