Dataset for "Designing Microservice Systems Using Patterns: An Empirical Study on Quality Trade-Offs"

This dataset provides materials used and produced in the context of the research study leading to the article Designing Microservice Systems Using Patterns: An Empirical Study on Quality Trade-Offs. It includes materials used to conduct the study, as well as aggregated and anonymized data produced in its context.

We investigated how practitioners perceive the impact of 14 patterns on 7 quality attributes. In particular, we conducted 9 semi-structured interviews to collect industry expertise regarding (1) knowledge and adoption of software patterns, (2) the perceived architectural trade-offs of patterns, and (3) metrics professionals use to measure quality attributes.

Research Objective

Our objective with this work was to obtain insights on the relevance of design patterns in industry, how practitioners perceive their influence on software qualities as a consequence of their usage, and what metrics practitioners use, if any, to determine these derived effects, reflected as software qualities.

Research Questions

• RQ1: What is the rationale for the adoption of patterns in microservices systems?
• RQ2: How are QAs influenced as a result of applying microservice patterns?
• RQ3: How are QAs measured in microservices?

Interview Artifacts and Results

Materials used to conduct the study:

• interview-signup-form.pdf - Form used by participants to signup for the study. The data collected through this form was used to characterize the participant, and make an early assessment of her or his experience with software development and microservices.
• interview-guide.pdf - Notes used by the researcher to conduct the interviews, including the general structure to follow.
• interview-helper.pdf - Slides used by the researcher during the interviews, to illustrate each of the patterns.

Data produced in the context of the study:

• PreliminaryTradeoffAnalysis.csv - Benefits and liabilities gathered by the researchers from the description of the analyzed patterns. Those classified QAs as mixed when we considered that assigning a positive or negative value was a too simplistic judgement.
• Demographics.csv - Demographic data for each of the interviewees.
• InterviewsStats.csv - Simple statistics regarding the interviews, including number of words of the transcripts and duration of the interviews.
• Adoptions.csv - Reported patterns adopted by the interviewees.
• Gains.csv - Gains of each pattern as reported by interviewees.
• Pains.csv - Pains of each pattern as reported by interviewees.
• Indicators.csv - Reported indicators for measuring each QA.
• Techniques.csv - Reported techniques for addressing each QA.
• ThirdPartyMonitoringTools.csv - Reported third-party monitoring tools and the number of participants who use them.