BROWSE
Detail
Conference Paper
Self-adaptive SW Framework for Smart IoT Environments
Cited - time in 
Share 
Share
- Authors
- Issue Date
- 2016-06
- Citation
- Conference on Network Softwarization (NetSoft) 2016, pp.1-1
- Language
- English
- Type
- Conference Paper
- Abstract
- With the rapid development of IoT (Internet of Things) [1] technologies, IoT devices have a lot different hardware and software capabilities. Most of typical IoT applications, however, do not utilize relevant capabilities of the IoT devices so that these applications may use more system resources (e.g. the number of CPU cores, allocated memory, and network bandwidth) than necessary. This causes a lot of resources wasted. This paper proposes a self-adaptive software framework [2, 3] which can guarantee optimal QoS for each IoT application by using static/dynamic information regarding IoT applications and system resources, where QoS means performance count, fps (frame per sec), etc. To achieve optimal QoS, the framework supports to manage and control IoT applications and system resources during application’s life cycle in real-time and dynamic way. The below figure shows the self-adaptive software framework composed of four managers. As shown in this figure, the framework covers two kinds of IoT applications: external self-adaptive application and internal self-adaptive application. Also, the framework supports a light-weight monitoring (CPU utilization is about 2%) and is a scalable middleware (a well-turned interfaces and information). Therefore, this framework can realize smart IoT services (e.g. smart glasses, smart mobile, smart home, smart car, etc) by keeping interoperability with other IoT devices.
- KSP Keywords
- CPU Utilization, Dynamic information, Hardware and software, IoT Applications, IoT devices, IoT environment, IoT services, Life cycle, Light-weight, Network bandwidth, Rapid development