Cloud-based Distributed Internet Measurement Platform
Perustieteiden korkeakoulu | Master's thesis
Unless otherwise stated, all rights belong to the author. You may download, display and print this publication for Your own personal use. Commercial use is prohibited.
Security and Cloud Computing
Master’s Programme in Security and Cloud Computing (SECCLO)
58 + 26
AbstractCloud computing has become an essential component of most networked applications today. Thus, Internet failures that prevent access to cloud data centres are sources of service disruptions and monetary losses. To prevent such failures, Internet measurement platforms are utilised to run network diagnostic tests to probe the Internet and identify anomalies. However, maintaining such platforms is complex, as it may involve large amounts of hardware and servers and has non-negligible monetary costs. This thesis investigates the possibility of implementing an Internet measurement platform leveraging the serverless computing paradigm. To this end, the work proposes the design and implementation of a novel cloud-based measurement platform in a serverless fashion. The proposed solution is then evaluated against the more traditional approach of using virtual machines in Google Compute Engine. Cost comparison between the two approaches shows that the serverless solution significantly reduces costs if the duration of the network measurement is below 22 minutes. As a drawback, performing measurements on the serverless approach exhibits slightly higher, yet negligible for measurement platforms, response time than the virtual machine-based approach; about 13.37% longer with 20.3 ms of difference.
SupervisorDi Francesco, Mario
Thesis advisorCorneo, Lorenzo
public internet, network measurement, cloud computing, serverless