Resource Adaptive Cloud Services for the Connected Vehicle A study of auto-scaling cloud for remote diagnosis of vehicles

Examensarbete för masterexamen

Please use this identifier to cite or link to this item: https://hdl.handle.net/20.500.12380/246542
Download file(s):
File Description SizeFormat 
246542.pdfFulltext1.27 MBAdobe PDFView/Open
Type: Examensarbete för masterexamen
Master Thesis
Title: Resource Adaptive Cloud Services for the Connected Vehicle A study of auto-scaling cloud for remote diagnosis of vehicles
Authors: Betselot, Hailu Abebe
Gebrecherkos, Alemayoh Haylay
Abstract: Cloud-based services are becoming more and more common thanks to the evergrowing cellular networks technology. There are several research works that aim at integrating cloud computing with different industries. The automotive industry is one of those areas that can benefit a lot from connectivity and cloud computing. Vehicular systems and cloud computing can be integrated to provide a safer and better driving experience. One of the areas that connected vehicles can benefit from cloud computing is remote diagnosis. Modern vehicles are made up of thousands of computing devices that work together. Troubleshooting and diagnosing these complex systems is not a trivial task as there are lots of components to diagnose and monitor. Currently, there are modern vehicles that store diagnostic data on a local hard drive and manufacturers use that offline diagnostic data to troubleshoot failures. Connected vehicles can leverage the increasingly fast and cheap mobile wireless networks to log diagnostic data so that it can be accessed and used my the manufacturers on fly. Cloud services can help manufacturers diagnose and even maintain vehicles remotely without recalling them to automotive workshops. They can monitor the different components while they are working and examine how the components are operating. This help diagnose problems that are not even detectable by the drivers. In order to realize the concept of connected vehicles, there is a need to have a robust cloud back-end system that adapts to data traffic from these vehicles. Vehicular data traffic is tends to fluctuate from place to place due to the mobile nature of the vehicles. Cloud instances that run in different geographical areas should be able to adapt to this changing nature of vehicular data traffic. Vehicular cloud service, like the other cloud services, it faces a number of challenges that need to be addressed such as privacy, security, scalability and a lack of standards. This thesis work aims at studying one of the core challenges: scalability. There is a need to have an architecture which can accommodate scaling number of users (connected vehicles, vehicle technicians) and services. There is extensive research work on autoscaling cloud systems that are not necessarily in the domain of connected vehicles. This work aims at taking advantage of these works to the area of vehicular cloud services. This is done by first building a prototype cloud back-end service and using it as a testbed to study autoscaling. This work proposes a cloud back-end service design and implementation that is capable of communicating with vehicles and provide access to manufacturers. We also looked into existing algorithmic implementations that enable vehicular cloud back-end services to adjust their resource usage according to the encountered traffic loads. Stress tests enabled us to preliminarily evaluate the implemented algorithm’s usability in the case of connected vehicles.
Keywords: Data- och informationsvetenskap;Computer and Information Science
Issue Date: 2016
Publisher: Chalmers tekniska högskola / Institutionen för data- och informationsteknik (Chalmers)
Chalmers University of Technology / Department of Computer Science and Engineering (Chalmers)
URI: https://hdl.handle.net/20.500.12380/246542
Collection:Examensarbeten för masterexamen // Master Theses



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.