Browsing by Author "de la Bastida, David"

Now showing 1 - 4 of 4

Extending IoT/M2M Scalability by Network Slicing
(2018-01-01)
de la Bastida, David
;
Lin, Fuchun Joseph
;
資訊工程學系
;
電機資訊國際碩士學位學程
;
Department of Computer Science
In this research, we have extended our initial effort in cloud-based IoT/M2M system scalability and developed a more robust solution by considering diverse QoS requirements from various IoT/M2M traffic patterns. Though our initial effort created a highly scalable architecture for IoT/M2M platforms based on OpenStack, it treated all IoT/M2M traffic without any discrepancy in the same underlying network (i.e. in the same network slice). Now, by leveraging software-defined networking in OpenStack and by using our traffic-slice optimal matching algorithm, we can direct different types of IoT traffic to feasible network slices in terms of QoS. Our experiments show that when compared with a system without network slicing, our scalability system performs better in terms of response time, power consumption, and computational cost.
High Scalability for Cloud-Based IoT/M2M Systems
(2016)
Cerritos, Eduardo
;
Lin, Fuchun Joseph
;
de la Bastida, David
;
資訊工程學系
;
電機資訊國際碩士學位學程
International standards such as ETSI SmartM2M and oneM2M have defined a client-server model as the common service architecture for IoT/M2M systems. Moreover, the IoT/M2M servers are normally provisioned in a cloud environment. This paper proposes a highly scalable architecture for all such cloud-based IoT/M2M systems. The uniqueness of our architecture is the incorporation of a Master Node in the cloud that is aware of system resources and incoming traffic so that it not only can dynamically decide load balancing policies but also proactively react to scalability needs. When compared with other cloud implementations without such a Master Node, our proposed system achieves faster response time and lower energy consumption.
OpenStack-based Highly Scalable IoT/M2M Platforms
(2017-01-01)
de la Bastida, David
;
Lin, Fuchun Joseph
;
資訊工程學系
;
電機資訊國際碩士學位學程
;
Department of Computer Science
The IoT/M2M platforms are emerging to provide a common service layer for various types of new IoT applications. It is foreseen billions of IoT/M2M devices will soon be connected to those IoT/M2M platforms hosted in the cloud. In this research we propose a highly scalable OpenStack-based architecture for IoT/M2M platforms. Our unique contribution consists of leveraging the advanced functionalities of OpenStack to introduce a master node specifically designed for IoT/M2M scalability in the cloud and a load balancing queue node cooperating with the master node for fair distribution of incoming traffic among platform nodes. Using four different types of IoT/M2M applications: smart meter, Bluetooth tags, eHealth, and video, we show that compared to LBAAS and Heat that are OpenStack native load balancing and scalability functions, in most cases our proposed system is capable of achieving the fastest response time and the lowest computational cost with acceptable tradeoff of more energy consumption.
Performance Evaluation of Open5GCore over KVM and Docker by Using Open5GMTC
(2018-01-01)
Chang, Hung-Cheng
;
Qiu, Bo-Jun
;
Chiu, Chen-Hao
;
Chen, Jyh-Cheng
;
Lin, Fuchun Joseph
Virtualizing the EPC (Evolved Packet Core) in modern carrier-grade networks provides more flexible and reliable network services by decoupling dependence on the hardware. Hypervisor and container are two popular virtualization techniques nowadays to enable virtualization of EPC. In this paper, we compared the Open5GCore, a 5G EPC prototype, over KVM and Docker, which are representatives of the hypervisor and container techniques. Moreover, to verify how each supports new services for Internet of Things (IoT) and Machine-to-Machine (M2M) communications, we also analyze the performance of two systems with the IoT and M2M traffic generated by Open5GMTC, a prototype tool for IoT/M2M platforms.