The Experts below are selected from a list of 35133 Experts worldwide ranked by ideXlab platform

Liu Cong - One of the best experts on this subject based on the ideXlab platform.

  • Co-Optimizing Performance and Memory FootprintVia Integrated CPU/GPU Memory Management, anImplementation on Autonomous Driving Platform
    2020
    Co-Authors: Bateni Soroush, Wang Zhendong, Zhu Yuankun, Hu Yang, Liu Cong
    Abstract:

    Cutting-edge embedded system applications, such as self-driving cars and unmanned drone software, are reliant on integrated CPU/GPU platforms for their DNNs-driven workload, such as perception and other highly parallel components. In this work, we set out to explore the hidden Performance Implication of GPU memory management methods of integrated CPU/GPU architecture. Through a series of experiments on micro-benchmarks and real-world workloads, we find that the Performance under different memory management methods may vary according to application characteristics. Based on this observation, we develop a Performance model that can predict system overhead for each memory management method based on application characteristics. Guided by the Performance model, we further propose a runtime scheduler. By conducting per-task memory management policy switching and kernel overlapping, the scheduler can significantly relieve the system memory pressure and reduce the multitasking co-run response time. We have implemented and extensively evaluated our system prototype on the NVIDIA Jetson TX2, Drive PX2, and Xavier AGX platforms, using both Rodinia benchmark suite and two real-world case studies of drone software and autonomous driving software

Cong Liu - One of the best experts on this subject based on the ideXlab platform.

  • co optimizing Performance and memory footprint via integrated cpu gpu memory management an implementation on autonomous driving platform
    Real Time Technology and Applications Symposium, 2020
    Co-Authors: Soroush Bateni, Zhendong Wang, Yuankun Zhu, Cong Liu
    Abstract:

    Cutting-edge embedded system applications, such as self-driving cars and unmanned drone software, are reliant on integrated CPU/GPU platforms for their DNNs-driven workload, such as perception and other highly parallel components. In this work, we set out to explore the hidden Performance Implication of GPU memory management methods of integrated CPU/GPU architecture. Through a series of experiments on micro-benchmarks and real-world workloads, we find that the Performance under different memory management methods may vary according to application characteristics. Based on this observation, we develop a Performance model that can predict system overhead for each memory management method based on application characteristics. Guided by the Performance model, we further propose a runtime scheduler. By conducting per-task memory management policy switching and kernel overlapping, the scheduler can significantly relieve the system memory pressure and reduce the multitasking co-run response time. We have implemented and extensively evaluated our system prototype on the NVIDIA Jetson TX2, Drive PX2, and Xavier AGX platforms, using both Rodinia benchmark suite and two real-world case studies of drone software and autonomous driving software.

Bateni Soroush - One of the best experts on this subject based on the ideXlab platform.

  • Co-Optimizing Performance and Memory FootprintVia Integrated CPU/GPU Memory Management, anImplementation on Autonomous Driving Platform
    2020
    Co-Authors: Bateni Soroush, Wang Zhendong, Zhu Yuankun, Hu Yang, Liu Cong
    Abstract:

    Cutting-edge embedded system applications, such as self-driving cars and unmanned drone software, are reliant on integrated CPU/GPU platforms for their DNNs-driven workload, such as perception and other highly parallel components. In this work, we set out to explore the hidden Performance Implication of GPU memory management methods of integrated CPU/GPU architecture. Through a series of experiments on micro-benchmarks and real-world workloads, we find that the Performance under different memory management methods may vary according to application characteristics. Based on this observation, we develop a Performance model that can predict system overhead for each memory management method based on application characteristics. Guided by the Performance model, we further propose a runtime scheduler. By conducting per-task memory management policy switching and kernel overlapping, the scheduler can significantly relieve the system memory pressure and reduce the multitasking co-run response time. We have implemented and extensively evaluated our system prototype on the NVIDIA Jetson TX2, Drive PX2, and Xavier AGX platforms, using both Rodinia benchmark suite and two real-world case studies of drone software and autonomous driving software

Soroush Bateni - One of the best experts on this subject based on the ideXlab platform.

  • co optimizing Performance and memory footprint via integrated cpu gpu memory management an implementation on autonomous driving platform
    Real Time Technology and Applications Symposium, 2020
    Co-Authors: Soroush Bateni, Zhendong Wang, Yuankun Zhu, Cong Liu
    Abstract:

    Cutting-edge embedded system applications, such as self-driving cars and unmanned drone software, are reliant on integrated CPU/GPU platforms for their DNNs-driven workload, such as perception and other highly parallel components. In this work, we set out to explore the hidden Performance Implication of GPU memory management methods of integrated CPU/GPU architecture. Through a series of experiments on micro-benchmarks and real-world workloads, we find that the Performance under different memory management methods may vary according to application characteristics. Based on this observation, we develop a Performance model that can predict system overhead for each memory management method based on application characteristics. Guided by the Performance model, we further propose a runtime scheduler. By conducting per-task memory management policy switching and kernel overlapping, the scheduler can significantly relieve the system memory pressure and reduce the multitasking co-run response time. We have implemented and extensively evaluated our system prototype on the NVIDIA Jetson TX2, Drive PX2, and Xavier AGX platforms, using both Rodinia benchmark suite and two real-world case studies of drone software and autonomous driving software.

George Lee - One of the best experts on this subject based on the ideXlab platform.

  • the Performance Implication of goal achievability in incentive contracts and feedback
    Social Science Research Network, 2013
    Co-Authors: Yasheng Chen, Johnny Jermias, George Lee
    Abstract:

    This study investigates the Performance feedback and goal achievability in incentive effects on employees’ effort and Performance. We perform an experiment to examine whether the use of goal-specific feedback and incentive contracts have an interaction effect on task Performance. Using the Mirametrix S2 eye tracking device to measure the level of effort, we find that the feedback effect on effort depend on goal achievability specified in the incentive contract. Specifically, we find that when employees are contracted based on achievable goals, feedback decreases their level of effort. By contrast, when employees are contracted based on more challenging but attainable goals, feedback increases their level of effort. Furthermore, we find that the level of effort has a significant positive impact on task Performance. These findings have important Implications for the design of control and compensation systems in organizations that aim for a higher employees’ Performance.