Next Article in Journal
Energy Enhancement and Energy Spread Compression of Electron Beams in a Hybrid Laser-Plasma Wakefield Accelerator
Next Article in Special Issue
Latency-Classification-Based Deadline-Aware Task Offloading Algorithm in Mobile Edge Computing Environments
Previous Article in Journal
Efficient Weights Quantization of Convolutional Neural Networks Using Kernel Density Estimation based Non-uniform Quantizer
Previous Article in Special Issue
A Smart System for Text-Lifelog Generation from Wearable Cameras in Smart Environment Using Concept-Augmented Image Captioning with Modified Beam Search Strategy
Open AccessArticle

EDCrammer: An Efficient Caching Rate-Control Algorithm for Streaming Data on Resource-Limited Edge Nodes

Department of Computer Science and Engineering, Kyung Hee University, 1732, Deogyeong-daero, Giheung-gu, Yongin-si, Gyeonggi-do 17104, Korea
*
Author to whom correspondence should be addressed.
Appl. Sci. 2019, 9(12), 2560; https://doi.org/10.3390/app9122560
Received: 3 May 2019 / Revised: 14 June 2019 / Accepted: 20 June 2019 / Published: 23 June 2019
(This article belongs to the Special Issue Edge Computing Applications in IoT)
This paper explores data caching as a key factor of edge computing. State-of-the-art research of data caching on edge nodes mainly considers reactive and proactive caching, and machine learning based caching, which could be a heavy task for edge nodes. However, edge nodes usually have relatively lower computing resources than cloud datacenters as those are geo-distributed from the administrator. Therefore, a caching algorithm should be lightweight for saving computing resources on edge nodes. In addition, the data caching should be agile because it has to support high-quality services on edge nodes. Accordingly, this paper proposes a lightweight, agile caching algorithm, EDCrammer (Efficient Data Crammer), which performs agile operations to control caching rate for streaming data by using the enhanced PID (Proportional-Integral-Differential) controller. Experimental results using this lightweight, agile caching algorithm show its significant value in each scenario. In four common scenarios, the desired cache utilization was reached in 1.1 s on average and then maintained within a 4–7% deviation. The cache hit ratio is about 96%, and the optimal cache capacity is around 1.5 MB. Thus, EDCrammer can help distribute the streaming data traffic to the edge nodes, mitigate the uplink load on the central cloud, and ultimately provide users with high-quality video services. We also hope that EDCrammer can improve overall service quality in 5G environment, Augmented Reality/Virtual Reality (AR/VR), Intelligent Transportation System (ITS), Internet of Things (IoT), etc. View Full-Text
Keywords: resource-limited edge node; lightweight and agile caching; advanced PID controller; qualified caching-as-a-service resource-limited edge node; lightweight and agile caching; advanced PID controller; qualified caching-as-a-service
Show Figures

Figure 1

MDPI and ACS Style

Kim, Y.; Huh, E.-N. EDCrammer: An Efficient Caching Rate-Control Algorithm for Streaming Data on Resource-Limited Edge Nodes. Appl. Sci. 2019, 9, 2560.

Show more citation formats Show less citations formats
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Access Map by Country/Region

1
Search more from Scilit
 
Search
Back to TopTop