Dataset Information

Energy efficient service placement in fog computing.

ABSTRACT: The Internet of Things (IoT) concept evolved into a slew of applications. To satisfy the requests of these applications, using cloud computing is troublesome because of the high latency caused by the distance between IoT devices and cloud resources. Fog computing has become promising with its geographically distributed infrastructure for providing resources using fog nodes near IoT devices, thereby reducing the bandwidth and latency. A geographical distribution, heterogeneity and resource constraints of fog nodes introduce the key challenge of placing application modules/services in such a large scale infrastructure. In this work, we propose an improved version of the JAYA approach for optimal placement of modules that minimizes the energy consumption of a fog landscape. We analyzed the performance in terms of energy consumption, network usage, delays and execution time. Using iFogSim, we ran simulations and observed that our approach reduces on average 31% of the energy consumption compared to modern methods.

SUBMITTER: Vadde U

PROVIDER: S-EPMC9455019 | biostudies-literature | 2022

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Energy efficient service placement in fog computing.

Vadde Usha U Kompalli Vijaya Sri VS

PeerJ. Computer science 20220719

The Internet of Things (IoT) concept evolved into a slew of applications. To satisfy the requests of these applications, using cloud computing is troublesome because of the high latency caused by the distance between IoT devices and cloud resources. Fog computing has become promising with its geographically distributed infrastructure for providing resources using fog nodes near IoT devices, thereby reducing the bandwidth and latency. A geographical distribution, heterogeneity and resource constr ...[more]

PMID: 36092002

Similar Datasets

Project description:The latest Internet of Things (IoT) edge-centric architectures allow for unburdening higher layers from part of their computational and data processing requirements. In the specific case of fog computing systems, they reduce greatly the requirements of cloud-centric systems by processing in fog gateways part of the data generated by end devices, thus providing services that were previously offered by a remote cloud. Thanks to recent advances in System-on-Chip (SoC) energy efficiency, it is currently possible to create IoT end devices with enough computational power to process the data generated by their sensors and actuators while providing complex services, which in recent years derived into the development of the mist computing paradigm. To allow mist computing nodes to provide the previously mentioned benefits and guarantee the same level of security as in other architectures, end-to-end standard security mechanisms need to be implemented. In this paper, a high-security energy-efficient fog and mist computing architecture and a testbed are presented and evaluated. The testbed makes use of Transport Layer Security (TLS) 1.2 Elliptic Curve Cryptography (ECC) and Rivest-Shamir-Adleman (RSA) cipher suites (that comply with the yet to come TLS 1.3 standard requirements), which are evaluated and compared in terms of energy consumption and data throughput for a fog gateway and two mist end devices. The obtained results allow a conclusion that ECC outperforms RSA in both energy consumption and data throughput for all the tested security levels. Moreover, the importance of selecting a proper ECC curve is demonstrated, showing that, for the tested devices, some curves present worse energy consumption and data throughput than other curves that provide a higher security level. As a result, this article not only presents a novel mist computing testbed, but also provides guidelines for future researchers to find out efficient and secure implementations for advanced IoT devices.

Dataset Information

Energy efficient service placement in fog computing.

Publications

Energy efficient service placement in fog computing.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets