Internet of Things (IoT) applications have seen exponential growth in the past several years. High throughput, wide bandwidth, Large number of connected devices, and high reliability in such applications are highly de...
详细信息
Cache partitioning is a technique to reduce interference among tasks accessing the shared caches. To make this technique effective, cache segments must be given to the tasks that can benefit most from having their dat...
详细信息
ISBN:
(纸本)9781450399838
Cache partitioning is a technique to reduce interference among tasks accessing the shared caches. To make this technique effective, cache segments must be given to the tasks that can benefit most from having their data and instructions cached for faster execution. The existing partitioning schemes for real-timesystems divide the available cache among the tasks to guarantee their schedulability which is the sole optimization criterion. However, it is also preferable, especially in systems with power constraints or mixed criticalities, to reduce the total cache usage for real-time tasks. In this paper, we develop optimization algorithms for cache partitioning that, besides ensuring schedulability, also minimize cache usage. We consider both preemptive and non-preemptive scheduling policies on single-processor systems. For preemptive scheduling, we formulate the problem as an integer quadratically constrained program and propose an efficient heuristic achieving near-optimal solutions. For non-preemptive scheduling, we combine linear and binary search techniques with different schedulability tests. Our experiments based on synthetic task sets with parameters from real-world embeddedapplications show that the proposed heuristic: (i) achieves an average optimality gap of 0.79% within 0.1x run time of a mathematical programming solver and (ii) reduces average cache usage by 39.15% compared to existing cache partitioning approaches. Besides, we find that for large task sets with high utilization, non-preemptive scheduling can use less cache than preemptive to guarantee schedulability.
The rapid development of intelligent machine learning and 5G/6G has led to smart terminal devices for various applications like auto-driving, AR, and smart farms. However, these applications have strict Quality of Ser...
详细信息
ISBN:
(纸本)9798350358261;9798350358278
The rapid development of intelligent machine learning and 5G/6G has led to smart terminal devices for various applications like auto-driving, AR, and smart farms. However, these applications have strict Quality of Service (QoS) requirements, exceeding the capabilities of mobile devices due to limited resources. Mobile edge computing offers a promising solution by enabling task offloading to nearby edge servers with better computing capabilities and a stable energy supply. For mobile edge systems, minimizing overall cost while ensuring QoS for mobile users is a challenging problem to solve effectively. In this paper, we first formulate the performance model for mobile edge systems with real data from deep learning scenarios. Then, we formulate the resource management in mobile edge system problem as a non-convex fractional programming problem with multiple coupled variables to minimize the latency and energy consumption while meeting QoS requirements. To solve the non-convex problem, a novel fractional programming technique is proposed to decouple the variables by considering fractional transformation to greatly reduce the complexity. Then we achieve the jointly optimal solution of the CPU frequency and offloading strategy with successive convex approximation, and Karush-Kuhn-Tucker (KKT) conditions. The experimental results show that our proposed algorithm can achieve cost-effective solutions while meeting QoS requirements over baselines.
Heightened awareness of the impact of climate change has led to the rapidly increasing penetration of renewable energy resources in electric energy distribution systems. Those distributed energy resources (DERs), most...
详细信息
ISBN:
(纸本)9798331531768;9798331531751
Heightened awareness of the impact of climate change has led to the rapidly increasing penetration of renewable energy resources in electric energy distribution systems. Those distributed energy resources (DERs), mostly inverter-based, can act as resilience sources for the grid but also introduce new control, stability, and cybersecurity challenges. This work proposes a digital twin (DT) that combines a real-time electromagnetic transient power system simulation and a practical model for communication network simulation. The digital twin allows the testing of novel control and cybersecurity strategies, including machine learning-based anomaly detection. Simulations using the Virginia Tech Electric Service (VTES) as a test case demonstrate the capability of adequately controlled resources, including solar PV, energy storage, and a synchronous generator, to enhance resilience by providing energy to critical loads. The DERs comply with ieee disturbance ride-through requirements, and switching transients are maintained within acceptable limits. A comprehensive DER-based resilience plan is developed and validated for the Virginia Tech Smart Grid.
The increasing proliferation of cyber-physical systems in a multitude of applications presents a pressing need for effective methods of securing such devices. Many such systems are subject to tight timing constraints,...
详细信息
ISBN:
(纸本)9781665409674
The increasing proliferation of cyber-physical systems in a multitude of applications presents a pressing need for effective methods of securing such devices. Many such systems are subject to tight timing constraints, which are poorly suited to traditional security methods due to the large runtime overhead and execution time variation introduced. However, the regular (and well documented) timing specifications of real-timesystems open up new avenues with which such systems can be secured. This paper contributes T-SYS, a timed-system method of detecting intrusions into real-timesystems via timing anomalies. A prototype implementation of T-SYS is integrated into a commercial real-time operating system (RTOS) in order to demonstrate its feasibility. Further, a compiler-based tool is developed to realize a T-SYS implementation with elastic timing bounds. This tool supports integration of T-SYS protection into applications as well as the RTOS the kernel itself. Results on an ARM hardware platform with benchmark tasks including those drawn from an open-source UAV code base compare T-SYS with another method of timing-based intrusion detection and assess its effectiveness in terms of detecting attacks as they intrude a system.
real-timeapplications of Music Information Retrieval (MIR) have been gaining interest as of recently. However, as deep learning becomes more and more ubiquitous for music analysis tasks, several challenges and limita...
详细信息
With the advent of the big data era, Data Stream Management systems (DSMS) have gradually become important tools for processing massive real-time data. This paper studies the data mining technology within DSMS, explor...
详细信息
This paper examines how data analytics and the Internet of Things (IoT) enhance smart city functionalities, focusing on urban infrastructure management, including energy consumption, traffic systems, and public safety...
详细信息
With the continuous development of smart city systems and architectures and their real-world implementation arises the need for a reliable way of verifying their integrity. Most smart city set-ups consist of subsystem...
详细信息
The paper presents the development and implementation of advanced algorithms for zigzag and contour parallel hatching techniques used in industrial applications such as laser marking. The primary objective is to enhan...
详细信息
暂无评论