Modeling, Analysis, and Hard Real-time Scheduling of Adaptive Streaming Applications
Jiali Teddy Zhai, Sobhan Niknam, Todor Stefanov
备注：
Accepted for presentation at EMSOFT 2018 and for publication in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems (TCAD) as part of the ESWEEK-TCAD special issue
In real-time systems, the application's behavior has to be predictable at compile-time to guarantee timing constraints. However, modern streaming applications which exhibit adaptive behavior due to mode switching at run-time, may degrade system predictability due to unknown behavior of the application during mode transitions. Therefore, proper temporal analysis during mode transitions is imperative to preserve system predictability. To this end, in this paper, we initially introduce Mode Aware Data Flow (MADF) which is our new predictable Model of Computation (MoC) to efficiently capture the behavior of adaptive streaming applications. Then, as an important part of the operational semantics of MADF, we propose the Maximum-Overlap Offset (MOO) which is our novel protocol for mode transitions. The main advantage of this transition protocol is that, in contrast to self-timed transition protocols, it avoids timing interference between modes upon mode transitions. As a result, any mode transition can be analyzed independently from the mode transitions that occurred in the past. Based on this transition protocol, we propose a hard real-time analysis as well to guarantee timing constraints by avoiding processor overloading during mode transitions. Therefore, using this protocol, we can derive a lower bound and an upper bound on the earliest starting time of the tasks in the new mode during mode transitions in such a way that hard real-time constraints are respected.
Hardware Architecture Proposal for TEDA algorithm to Data Streaming Anomaly Detection
Federal University of Rio Grande do Norte（里约热内卢联邦大学？不对，等下，Rio Grande do Norte是北里奥格兰德，对，Federal University of Rio Grande do Norte是北里奥格兰德联邦大学）；Federal Institute of Education, Science and Technology of Rio Grande do Norte（北里奥格兰德联邦教育、科学与技术学院）；University of Granada（格拉纳达大学）；Harvard University（哈佛大学）
RTGPU: Real-Time GPU Scheduling of Hard Deadline Parallel Tasks with Fine-Grain Utilization
IEEE（电气与电子工程师协会）；NVIDIA（英伟达）
许多新兴的网络物理系统，如自动驾驶汽车和机器人，严重依赖人工智能和机器学习算法来执行重要的系统操作。由于这些高度并行的应用程序是计算密集型的，它们需要由图形处理单元（GPU）加速以满足严格的时序约束。然而，尽管GPU被广泛采用，但在提供严格的实时的保证的同时有效地调度多个GPU应用仍然是一个挑战。在本文中，我们提出了RTGPU，它可以调度多个GPU应用程序的执行在实时的，以满足硬期限。每个GPU应用程序可以有多个CPU执行和内存复制段，以及GPU内核。我们从一个模型开始，明确说明这些应用程序的CPU和内存复制段。然后，我们考虑GPU架构中的GPU内核的精确定时模型的发展，并利用称为持久线程的技术来实现细粒度的内核调度，通过交错执行提高性能。接下来，我们提出了一种通用的方法，用于实时的调度并行GPU应用程序。最后，调度多个并行GPU应用程序，我们提出了一个实用实时的调度算法的基础上联合调度和网格搜索（GPU内核段）与单处理器固定优先级调度（多CPU和内存复制段）。通过在一个真实的NVIDIA GTX 1080 Ti GPU系统上的全面验证和评估，我们的方法提供了比以前工作更好的可扩展性，并为满足多个GPU应用的硬性期限提供了实时的保证。
