English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Many emerging cyber-physical systems, such as autonomous vehicles and robots,rely heavily on artificial intelligence and machine learning algorithms toperform important system operations. Since these highly parallel applicationsare computationally intensive, they need to be accelerated by graphicsprocessing units (GPUs) to meet stringent timing constraints. However, despitethe wide adoption of GPUs, efficiently scheduling multiple GPU applicationswhile providing rigorous real-time guarantees remains a challenge. In thispaper, we propose RTGPU, which can schedule the execution of multiple GPUapplications in real-time to meet hard deadlines. Each GPU application can havemultiple CPU execution and memory copy segments, as well as GPU kernels. Westart with a model to explicitly account for the CPU and memory copy segmentsof these applications. We then consider the GPU architecture in the developmentof a precise timing model for the GPU kernels and leverage a technique known aspersistent threads to implement fine-grained kernel scheduling with improvedperformance through interleaved execution. Next, we propose a general methodfor scheduling parallel GPU applications in real time. Finally, to schedulemultiple parallel GPU applications, we propose a practical real-time schedulingalgorithm based on federated scheduling and grid search (for GPU kernelsegments) with uniprocessor fixed priority scheduling (for multiple CPU andmemory copy segments). Our approach provides superior schedulability comparedwith previous work, and gives real-time guarantees to meet hard deadlines formultiple GPU applications according to comprehensive validation and evaluationon a real NVIDIA GTX1080Ti GPU system.

RTGPU: Real-Time GPU Scheduling of Hard Deadline Parallel Tasks with Fine-Grain Utilization