112 – Unchaining microservice chains

In modern software development, microservices are crucial. Instead of a monolithic application, where everything is tightly coupled, microservice architecture offers a way to develop, deploy and maintain services independently. This enhances agility, resilience, and scalability, but it also introduces challenges concerning for example resource allocation and performance optimization.

In his doctoral thesis in Computer Science, Michel Gokan Khan addresses such challenges in large scale microservice chains, specifically in cloud native computing. In our conversation, he explains some of the key contributions of his research. Two of them being the PerfSim, a tool designed by Michel that is a performance simulator for cloud native system, and NFV-Inspector, another tool designed by him to be able to profile and analyse microservices specifically in network functions virtualization (NFV) environments. Michel’s research is also able to show how machine learning can be used for optimizing microservice chains in cloud environments. By contributions such as these and more, Michel’s research helps to reveal the true potential of artificial intelligence in profiling, modelling, simulating, and most importantly optimizing the performance and cost of running microservice chains.

Michel Gokan Khan’s doctoral thesis can be downloaded from DiVA: Unchaining Microservice Chains: Machine Learning Driven Optimization in Cloud Native Systems

97 – Privacy and security of digital health technologies

During the early phase of the Corona pandemic public health actors rapidly deployed contact tracing applications for mobile phones. The apps aimed at curbing the spread of COVID-19, by aiding manual contact tracing. In Samuel Wairimu’s research in Computer Science he investigates the security and privacy risks of these apps and how patients can be harmed by those risks.

In his licentiate thesis, Samuel has used an experimental approach to assess the risks and combined it with a theoretical approach which goes on to assess what impacts and consequences a state-sponsored cyberattack could have to the healthcare sector, and to patients especially.

In our conversation, Samuel explains the results of his study and he argues that more emphasis needs to be put on information security in the healthcare sector, as it is prone for experiencing cyberattacks and the risks for patients can be severe. 

Samuel Wairimu’s licentiate thesis can be downloaded from DiVA: Privacy and Security Analysis : Assessing Risks and Harm to Patients

80 – Empowering users of online services

Using online services means ticking a lot of boxes in consent forms, but do we always know what we are saying yes to? In this podcast we are talking to Farzaneh Karegar, PhD in Computer Science. In her research she proposed, designed, and tested usable and legally compliant tools and solutions that can empower users to take control of their data when using online services. 

We talk about the trade-offs between convenience and privacy when it comes to making online choices; tools that both users and service providers can benefit from and consent form designs that motivate users to pay more attention to what they are disclosing and for what purposes. Farzaneh introduces us also to dark patterns which are very prominent in, for example, cookie consent banners and explains why they can be a pitfall also for policy designers with good intentions. 

Farzaneh Karegar’s doctoral thesis can be downloaded from DiVA: The Lord of Their Data Under the GDPR?: Empowering Users Through Usable Transparency, Intervenability, and Consent

51 – Fixing WiFi

Do you have high speed internet to your home but still experience  buffering or lag when watching a movie or playing an online game? These problems may be related to your WiFi network. Good thing then that researchers are working on to fixing WiFi! In Toke Høiland Jørgensen’s research he not only produces more knowledge on WiFi performance barriers, he also develops mathematical and technical solutions to them. In our conversation Toke explains how the open source solutions developed by him and his colleagues can benefit all us speed-craving internet users.

Toke Høiland-Jørgensen’s doctoral thesis can be downloaded from DiVA: Bufferbloat and Beyond : Removing Performance Barriers in Real-World Networks

48 – Human aspects of data privacy

Securing data privacy on the internet is a wicked problem. It is wicked in that technological, legal, and human aspects of privacy are entangled. In her licentiate thesis Ala Sarah Alaqra has focused on the human aspects of data privacy. By letting users test a privacy enhancing scheme – redactable signatures – designed by Ala and her colleagues, her research investigates user perceptions and opinions of data privacy in eHealth. In our conversation Ala let us know more of redactable signatures, and how it can help increase data privacy in eHealth and possibly in other sectors as well. Ala Sarah Alaqra’s licentiate thesis can be downloaded from DiVA: The Wicked Problem of Privacy : Design Challenge for Crypto-based Solutions

41 – Improving transparency in personal data processing

The European Union’s General Data Protection Regulation has been implemented to protect citizens’ data privacy by, for example, increasing control over their personal data. However, computerized systems and web services are not always effectively designed to give users the control they are legally entitled to in a usable way. In her thesis, Farzaneh Karegar, Ph. D. student in computer science, develops new solutions that enhance transparency and make it easier for users to give better-informed consent to service providers handling personal data. In our conversation Farzaneh lets us know more about the solutions, and why it is important to continue to work to improve the gap between legally-compliant and usable services. Farzaneh Karegar’s licentiate thesis can be retrieved from DiVA: Towards Improving Transparency, Intervenability, and Consent in HCI 

32 – Personal data privacy

In our digitalised world more and more of our personal information is registered in network computers and servers. Stakeholders handling personal information thus need to make sure their systems are secure and maintain the integrity of individuals. Automated privacy audits is one approach to ensure that stakeholders do in fact maintain the privacy of personal data. But as Jenni Reuben shows in her research, these audits can themselves subject to privacy risks. In our conversation Jenni, Ph.D. student in computer science, tells more about these risks and of the model she proposes to prevent illegitimate access to personal data. To read Jenni Reuben’s dissertation please follow this link: Privacy-aware Use of Accountability Evidence

19 – How to make mobile health systems secure and privacy-preserving

Information security and privacy are vital to achieve high quality healthcare services. In this podcast we dig deeper into this issue together with Leonardo Horn Iwaya who has studied how to make Mobile Health (mHealth) systems secure and privacy-preserving. He tells us about the ever-increasing tendency of using mobile devices (e.g., mobile phones, sensors, smartwatches) to support medical and public health and the problems that come along with it due to failures in addressing security and privacy issues. After listening to this podcast you will also know more about the design of security and privacy enhancing technologies for mHealth, specially in Mobile Data Collection Systems, in which traditional paper forms are replaced by a mobile application for collecting and reporting health-related data used for public programs such as health surveys and surveillance. If you want to know more you can read his licentiate thesis here: http://urn.kb.se/resolve?urn=urn:nbn:se:kau:diva-46982

 

16 – Debloating Internet Access Networks

Ever got delayed on your way to surf the web? Then this podcast might be something for you. Toke Høiland-Jørgensen talks about his thesis “On the Bleeding Edge: Debloating Internet Access Networks” in which he studied why and how internet access delays occur. This is increasingly important in times when more and more devices are connected to the internet and applications become more interactive. We need our network to respond reliably and without unnecessary delay. Tokes work focuses on one of those delays. They are called bufferbloats and are a result of excess queueing delay in network routers that occur along the path. Learn more about what it is, why it is a problem and what can be done to make it better.