Hybrid-LRU Caching for Optimizing Data Storage and Retrieval in Edge Computing-Based Wearable Sensors

Jia, Gangyong; Han, Guangjie; Xie, Hongtianchen; Du, Jian Hua

doi:10.1109/jiot.2018.2834533

Cited by 26 publications

(11 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The rapid development of edge computing is raising higher and higher requirements to data storage and computing capacity of edge computing system. Plenty of researches have been done focusing on data storage [17]- [20]. At the meantime, the computing power is also a bottle neck of system performance.…”

Section: Related Workmentioning

confidence: 99%

A Hardware and Software Task-Scheduling Framework Based on CPU+FPGA Heterogeneous Architecture in Edge Computing

et al. 2019

Self Cite

View full text Add to dashboard Cite

Real-time performance is the primary requirement for edge computing systems. However, with the surge in data volume and the growing demand for computing power, a computing framework consisting solely of CPUs is no longer competent. As a result, CPU+ heterogeneous architecture using accelerators to improve edge computing systems' computing capacity has received great attention. The type of accelerators determines the performance of the edge computing system largely. The accelerators include Graphics Processing Unit (GPU), Application Specific Integrated Circuit (ASIC) and Field Programmable Gate Array (FPGA). FPGAs with its reconfigurability and high energy efficiency are widely used in many edge computing scenarios. Nontheless, the performance depends also on the scheduling efficiency between software tasks on CPUs and hardware tasks on FPGAs. Unfortunately, the existing strategies have not fully exploited the differences between hardware and software tasks, thus resulting in low scheduling efficiency. This paper proposes a task scheduling framework on the Dynamic Partial Reconfiguration (DPR) platform. We take full account of the characteristics of task switching overhead and predictable execution time of hardware tasks in DPR, and reduce the number of task-switching times and active tasks in the system, thus improving the scheduling efficiency. We conduct a set of experiments on the Zynq platform to verify the proposed framework. Experimental results demonstrate that when the execution time of the accelerator exceeds the reconfiguration cost by an order of magnitude, the efficiencies of all the cases are more than 98%, and the efficiencies can reach 90%-98% in the same order of magnitude.INDEX TERMS FPGA, edge computing, task scheduling, heterogeneous system, dynamic partial reconfigurable system.

show abstract

Section: Related Workmentioning

confidence: 99%

A Hardware and Software Task-Scheduling Framework Based on CPU+FPGA Heterogeneous Architecture in Edge Computing

et al. 2019

Self Cite

View full text Add to dashboard Cite

show abstract

“…It makes good performance in both task completion time and load balancing of the cloud task scheduling. Then an adaptive ant colony optimization algorithm (SAACO) is proposed in [13] for the shortcomings of PACO proposed in [12], such as parameter selection and pheromone updating. The algorithm uses the particle swarm optimization algorithm to adaptively update the parameters in the ACO algorithm and also improves the pheromone mechanism, which has better performance than PACO in minimum task completion time and system load balance [25], [26].…”

Section: Related Workmentioning

confidence: 99%

MCTE: Minimizes Task Completion Time and Execution Cost to Optimize Scheduling Performance for Smart Grid Cloud

et al. 2019

Self Cite

View full text Add to dashboard Cite

Cloud computing is popular in nowadays for its convenient and cheap. Grid provides services of available everywhere, and shares everyone. Therefore, smart grid cloud is a good way to manage data for sharing with all power supply stations. Grid cloud task scheduling is one of the key technologies that affect resource allocation efficiency in cloud computing environment. The advantages and disadvantages of scheduling algorithms will directly affect the scheduling performance of both cloud computing and the stability of the entire system platform. Cloud task scheduling problem has been proved to be a NP-hard problem, The traditional task scheduling algorithm can no longer meet the actual needs of cloud task scheduling, but the Heuristic algorithm is an effective method to solve this problem. This paper studies and analyzes the application of heuristic algorithms in cloud task scheduling problems, and proposes a cloud task scheduling strategy to minimize the task completion time and execution cost (MCTE) for the smart grid cloud. Then, carry out mathematical modeling on the grid cloud task scheduling problem. The experimental results show MCTE is well for the smart grid cloud. INDEX TERMS Smart grid cloud, task scheduling, particle swarm optimization-genetic algorithm.

show abstract

“…Such as Abdelwahab et al [27] propose User-Level Online Offloading Framework (ULOOF), a lightweight and efficient framework for mobile computation offloading, which is equipped with a decision engine that minimizes remote execution overhead. Jia et al [28]- [31] propose a series of new caching strategies such as Hybrid-LRU, cost aware cache replacement policy (CACRP), and dynamic adaptive replacement policy (DARP) to improve memory resources management performance. However, these previous studies ignore thread behaviors and lack target optimizations.…”

Section: Related Workmentioning

confidence: 99%

A Thread-Oriented Memory Resource Management Framework for Mobile Edge Computing

Zhu

et al. 2019

IEEE Access

Self Cite

View full text Add to dashboard Cite

Increasing the number of cores is one of the most effective methods to enhance performance. However, an extensive experimental study on mobile edge computing (e.g., Android devices) indicates that the memory management system has gradually become a key performance bottleneck. Studies on improving memory management mainly focus on exploring the trade-off between avoiding fragmentation and improving allocation efficiency. From our previous research, we know that the fragmentation is no longer a crucial bottleneck; instead, inter-and intra-thread behavior should be focused on, and thus, we introduce memory management based on thread behaviors (MMBTB). Unfortunately, it lacks a unified optimization program interface and good architecture. Consequently, in this paper, we propose a memory resource management at operating system (OS) layer of mobile edge computing, called the thread-oriented memory management layer (TOMML) to address this problem, which follows the microkernel architecture pattern and can meet the user's requirements for selecting plug-ins to achieve different optimization goals. This paper is divided into several sections as follows. First, we demonstrate the efficiency of TOMML through theoretical simulation and experimentation. The experimental result is that TOMML can improve memory allocation efficiency by 12%-20%. Furthermore, we introduce a plug-in to save power, which can further promote 6%-25% bank free compared with previous excellent research.

show abstract

Hybrid-LRU Caching for Optimizing Data Storage and Retrieval in Edge Computing-Based Wearable Sensors

Cited by 26 publications

References 25 publications

A Hardware and Software Task-Scheduling Framework Based on CPU+FPGA Heterogeneous Architecture in Edge Computing

A Hardware and Software Task-Scheduling Framework Based on CPU+FPGA Heterogeneous Architecture in Edge Computing

MCTE: Minimizes Task Completion Time and Execution Cost to Optimize Scheduling Performance for Smart Grid Cloud

A Thread-Oriented Memory Resource Management Framework for Mobile Edge Computing

Contact Info

Product

Resources

About