Firework: Data Processing and Sharing for Hybrid Cloud-Edge Analytics

Zhang, Quan; Zhang, Qingyang; Shi, Weisong; Zhong, Hong

doi:10.1109/tpds.2018.2812177

Cited by 68 publications

(34 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, it provides an easy-to-use programming interface for both service providers and end users. The experimental results of [51] article demonstrate how Firework can reduce response latency and network bandwidth cost for a video analytics application compared to a cloud-centric solution.…”

Section: Data/video Analysis In the Edgementioning

confidence: 98%

“…Firework [51] provides distributed data processing and sharing via a virtual shared data view and service composition for Internet of Everything (IoE) applications. Firework collects and concatenates data and predefined functions from multiple stakeholders and data owners to a virtual shared data set with the care of data privacy protection.…”

Section: Data/video Analysis In the Edgementioning

confidence: 99%

See 1 more Smart Citation

5G support for Industrial IoT Applications— Challenges, Solutions, and Research gaps

Varga

Pető

Frankó

et al. 2020

Sensors

187

View full text Add to dashboard Cite

Industrial IoT has special communication requirements, including high reliability, low latency, flexibility, and security. These are instinctively provided by the 5G mobile technology, making it a successful candidate for supporting Industrial IoT (IIoT) scenarios. The aim of this paper is to identify current research challenges and solutions in relation to 5G-enabled Industrial IoT, based on the initial requirements and promises of both domains. The methodology of the paper follows the steps of surveying state-of-the art, comparing results to identify further challenges, and drawing conclusions as lessons learned for each research domain. These areas include IIoT applications and their requirements; mobile edge cloud; back-end performance tuning; network function virtualization; and security, blockchains for IIoT, Artificial Intelligence support for 5G, and private campus networks. Beside surveying the current challenges and solutions, the paper aims to provide meaningful comparisons for each of these areas (in relation to 5G-enabled IIoT) to draw conclusions on current research gaps.

show abstract

Section: Data/video Analysis In the Edgementioning

confidence: 98%

Section: Data/video Analysis In the Edgementioning

confidence: 99%

5G support for Industrial IoT Applications— Challenges, Solutions, and Research gaps

Varga

Pető

Frankó

et al. 2020

Sensors

187

View full text Add to dashboard Cite

show abstract

“…Edge computing model is promising for the killer applications with latency-sensitive requirements [22]. Edge computing-based video processing system can perform video data analytics at the edge node that is close to the cameras [23], [24]. Edge Video analytics can reduce video data transmission [25] on the network and improve the latency of video processing [26].…”

Section: Related Workmentioning

confidence: 99%

mVideo: Edge Computing Based Mobile Video Processing Systems

Sun

Sha

et al. 2020

IEEE Access

View full text Add to dashboard Cite

Computer vision is widely used to detect anomalies in video processing systems for public safety. Applying Deep Neural Networks (i.e., DNNs) in computer vision can achieve a high detection accuracy but it requires a huge amount of computing power, storage space, and video data. Thus, DNNs-based video analytics is mostly deployed in the cloud with video data steaming from a set of stationary cameras. There are mainly three issues in this setting. First, steaming a huge amount of video data from cameras to cloud leads to high bandwidth consumption and latency. Second, when DNNs are deployed on resource-limited devices like edge nodes to reduce communication costs, it is hard to achieve a high detection accuracy. Third, stationary cameras can only collect a limited amount of video data that covers a small area, so it barely satisfies the needs of the real-time analytics in applications like public safety. We propose a mobile edge computing-based video stream processing platform, mVideo, which conducts video analytics making full use of resources at the collaborative edge and cloud nodes. On the mVideo, a mechanism is designed to partition a video analysis task based on available resources on the mobile edge node. Then, the edge nodes pre-process video data using a lightweight DNN model and upload the results to cloud nodes for further analysis. Thus mVideo not only collects video data that covers a large area, but also reduces the communication costs. To validate the proposed platform, a face recognition application is deployed on the mVideo prototype. Experimental results reveal that compared with the existing cloud computing model, mVideo reduces video data volume transmitted to the cloud nodes and power consumption by up to 99.5% and 96.2%, respectively. mVideo also improves the execution time by 90.0% to optimize mobile video analytics performance. INDEX TERMS Mobile video analytics, edge computing, mobile cameras, public safety.

show abstract

“…For processing and monitoring, Zhang et al sent data captured by cameras to cloud storage [32]. This data transmission results in high response latency and bandwidth constraints, which in turn proves this solution to be inefficient and expensive.…”

Section: Literature Surveymentioning

confidence: 99%

An Inexpensive Upgradation of Legacy Cameras Using Software and Hardware Architecture for Monitoring and Tracking of Live Threats

Habiba

Awais

Jaleel³

2020

IEEE Access

View full text Add to dashboard Cite

Surveillance through digital cameras is increasing exponentially. A majority of these cameras are not smart cameras; therefore, they send their video stream to a central server where it is processed and analyzed for any threats. Typically, human operators or machine learning algorithms at the cloud analyzed and processed the post-event videos to track and locate the perpetrator or victim. The centralized approach leads to two primary shortcomings: 1) the high cost of cloud infrastructure; 2) lack of instant tracking and detection of the threat. One solution is to replace these legacy cameras with the smart cameras so they can process information locally. Although the solution is costly, it could solve the real-time threat detection issues. However, the need for a central server remains there, to construct the path of threat, when threat moves from one camera view to another. The existing distributed architectures for threat tracking shifts the load of threat capturing and processing from a central server to the edge nodes, which in turn reduces the computational power but does not remove the role of the central server completely. These architectures don't equip each camera of processing and communicating with each other. Further, in the existing distributed architectures, the local cameras are not able to store the path of the threat individually and just transmit the captured trajectory to the central body. This research proposed a second alternative that makes use of legacy cameras through additional hardware and software components such that they can process information and collaborate locally. The research addresses the challenge by introducing a low cost distributed threat tracking framework that allows the single camera to identify the threat and communicate its information to other cameras without involving the central server. The framework stores the information in a lightweight architecture that is inspired by the blockchain storage algorithm. The system also allows querying the path traveled by the threat at any stage. To evaluate the system, we performed two simulated experiments: one with a central server and another with the proposed distributed system. The results of the experiments showed that the time to track the threat through the proposed system was lower than the existing centralized system. Moreover, the proposed system predicted the paths of threats with an accuracy of 85.49%. In the future, the technique may be improved with reinforcement learning and other machine learning techniques. INDEX TERMS Video surveillance, smart cameras, distributed computing, software architecture, peer-topeer computing.

show abstract

Firework: Data Processing and Sharing for Hybrid Cloud-Edge Analytics

Cited by 68 publications

References 29 publications

5G support for Industrial IoT Applications— Challenges, Solutions, and Research gaps

5G support for Industrial IoT Applications— Challenges, Solutions, and Research gaps

mVideo: Edge Computing Based Mobile Video Processing Systems

An Inexpensive Upgradation of Legacy Cameras Using Software and Hardware Architecture for Monitoring and Tracking of Live Threats

Contact Info

Product

Resources

About