Titulo Estágio
ucXception fault injection framework
Áreas de especialidade
Engenharia de Software
Local do Estágio
DEI/CISUC
Enquadramento
The aim of this proposal is to develop and improve an existing framework for dependability evaluation. The framework, called ucXception, is an open-source project developed in our department and has been used in a variety of works to evaluate virtualized cloud computing systems, programming languages and REST middleware libraries. In fact, ucXception can be applied to any computer system and operates by executing fault injection campaigns where errors that emulate hardware faults and software bugs are inserted in the target system, thus creating a scenario where the application is affected by these faults and allowing the user to study how the system behaves.
This proposal is integrated in the European project VALU3S (https://valu3s.eu/) and the student selected will receive a BSc grant of 835 euros/month during 6 months, with the possibility of being renovated for additional 6 months.
Objetivo
This proposal has three main goals:
1) Improving the usability of the tool, so that a wider audience can use it even if they are not experts in the field. This step may imply developing user interfaces to configure the fault injection campaigns or to monitor the state of running campaigns.
2) Packaging ucXception using a well-known packaging or containerization technology. This step may imply the development of a Docker image containing the framework, which can then be made available so that anyone can easily and quickly set it up on their system.
3) Researching techniques for accelerating the process of fault injection. This step may consist in evaluating whether the injection of failures can produce results that are as accurate and representative as those obtained using fault injection. If this hypothesis is verified, then the usage of failure injection will lead to quicker campaigns and should be implemented into the framework.
This proposal aims to find one student that enjoys programming and is interested in learning more about dependability and fault injection. The proposal can be continued as a topic for a master thesis.
Plano de Trabalhos - Semestre 1
T1 (M1): State of the art review on fault injection and study of the existing code of ucXception.
T2 (M2): Analysis and identification of measures to improve current software fault injection, particularly concerning the usability of the technique/tool, automatization of the process, the accuracy of the injected faults.
T3 (M2-M4): Evaluation of different strategies and techniques to improve the software fault injection process.
T4 (M5): Writing the Intermediate report.
Plano de Trabalhos - Semestre 2
T5(M6): Design of the new ucXception 2.0, including rapid prototypes for the functionalities/user interface.
T6 M7-M9): Implementation of the ucXception 2.0, including the strategies researched in T3 such as packaging strategies.
T7 (M9-M10): Testing of the ucXception 2.0 tool and definition and execution of fault injection campaigns to validate the tool.
T8 (M10): Write a research paper to submit to an international conference.
T9 (M11): Write the MSc thesis.
Condições
The student selected for this proposal will have a BSc grant of 835 euros/month for 6 months. This grant can be renewed if the performance of the student is positive.
This MSc thesis proposal is integrated in the European project VALU3S (https://valu3s.eu/)
This MSc thesis will be advised by Frederico Cerveira and co-advised Henrique Madeira
Orientador
Frederico Cerveira
fmduarte@dei.uc.pt 📩