Henning Schulz's research while affiliated with Universität Stuttgart and other places

... Moreover, LWS explores different strategies of workload intensity modeling and combines them for workload simulation which can be intervened effectively for quick, extensible, and intervenable workload generation. Besides, some other approaches Fei et al., 2020;Erradi et al., 2021;Schulz et al., 2021;Feng et al., 2022) attempt to forecast future workloads based on time series forecasting and workload specification. Different from these approaches, LWS provides the e2e framework for workload simulation rather than specific forecasts or analysis for concrete systems. ...

... We have designed the proposed machine learning approach in the context of the architecture for cloud-edge orchestration of IoT applications [49], [50], [51]. As illustrated in Fig. 1, it is positioned between the Cloud layer and the Edge layer as it receives data streams from IoT devices situated in both the Edge and Mobile Edge layers. ...

... Test runs can often last many hours and often have to be run outside normal business times. Together, the time consuming and complex nature of NFR testing also adds to its cost [9]. ...

... Configuration Issues: Microservices systems can have a large number of services and their instances to configure and manage with third-party systems, deployment platforms, and log templates [51]. It is essential that microser-vices systems should have the ability to track and manage the code and configuration changes. ...

... WESSBAS (ögele et al., 2018) combines approaches based on CBMGs and EFSMs, extracts GaAs from production session logs automatically, and generates executable workload models. Schulz et al. (Schulz et al., 2019) and Barnert et al. (Barnert and Krcmar, 2021) extend WESSBAS to microservice applications and databases by tailoring logs and modifying the Markov chain respectively. Besides, the stochastic formoriented analysis (Draheim et al., 2006;Lutteroth and Weber, 2008), the probabilistic timed automata (Abbors et al., 2012), the context-based sequential action (Zhou et al., 2014) and the latent features computed by URI space mapping vectors (Erradi et al., 2021) are also applied in user behavior abstraction. ...

... Software engineering is a branch of computer science that includes the design, development, evaluation, and implementation of computer software [37,38]. System testing and maintenance remains a key activity in software engineering that is concerned with error reporting and correction [39][40][41][42]. Academic software engineering projects are increasingly gaining attention as a reliable method to assess competency levels among ICT students [4,14,19]. ...

... For our pilot study, we are utilizing a dataset of traces that was generated in a previous work of the second author [40]. The dataset was created using the widely-used microservices benchmark system TrainTicket [16,26,[44][45][46], and the load generator PPTAM [4]. Our technology stack includes the use of Jaeger [38] as the distributed tracing collector and Elasticsearch [11] for storing both common and optimized traces. ...

... We use the data obtained from the investigated system to illustrate our approach. The computation of the pass/fail criteria (performance/scalability requirement) uses the approach proposed by Avritzer et al. (2018), which calculates pass/fail criteria as: average + 3 · standard deviation of the no-load response time measurement. Then, these values are used to calculate the two metrics used to detect the SPA: slope and normalized distance. ...

... We considered the following methodologies:"Model-based" -if the authors emphasize modeling abstractions, "Empirical" -if the authors designed a model-free approach and their decision process is based on the collected static or runtime data, and "Hybrid/Other" -if the authors used a combination of modelbased and empirical approach or other methods. [10][11], [22][23][24], [27][28][29][30][31], [49] §II-III, §VII [10], [19], [32] [40], [45][46], [50] §II-IV [5][6][7], [20], [51] §II [8][9], [48] §II-III [25][26], [49] §II, §VII [16][17][18][19], [21] [ [34][35][36], [47], [50] §II, §V [12][13][14], [33], [37][38][39], [41][42][43][44] §III, §VI [14] §IV: [15] §V: [16]- [18] §II: [5]- [7], [19], [20] §V: [21] §VII: [22]- [24] §VII: [25], [26] §VII: [12], [27]- [31] Reliability / Fault-tolerance §III: [11], [32], [33] §V: [34]- [36] §II: [6] Safety / Correctness §VI: [37], [38] §VI: [39] §IV: [40] §VI: [41]- [44] Survivability / Security §IV: [45] §IV: [46] §V: [47] Cost of ownership §III: [10], [11] §II: [20] Any quality attribute §II: [48]- [50] §II: [51] ...

... For many years business process management (BPM) approach has been used to analyze business processes in many organizations. Traditional business process management methods assume the following five stages in the business process cycle: Process design consisting of creating a new or changing an existing process model; creating a process instance, its configuration and startup; executing; monitoring the execution of the process instance; and analyzing the performance of the process for improvement [1][2][3]. ...