Dr.-Ing. Florian Lautenschlager
Cooperative Exploration and Analysis of Software in a Virtual/Augmented Reality Appliance
(Third Party Funds Group – Sub project)Overall project: Cooperative Exploration and Analysis of Software in a Virtual/Augmented Reality Appliance
Term: 01.09.2018 - 31.12.2022
Funding source: Bundesministerium für Wirtschaft und Technologie (BMWi)
URL: https://www2.cs.fau.de/research/Holoware/Understanding software has a large share in the programming efforts of a software systems, up to 30% in development projects and up to 80% in maintenance projects. Therefore, an efficient and effective way for comprehending software is necessary in a modern software engineering workplace. Three-dimensional software visualization already boosts comprehension and efficiency, so utilization of the latest virtual reality techniques seems natural. Within the scope of the Holoware project, we create an environment to cooperatively explore and analyze a software project using virtual/augmented reality techniques as well as artificial intelligence algorithms. The software project in question is being visualized in said virtual reality, such that multiple participants can simultaneously explore and analyze the software. They can cooperate by communicating about their findings. Different participants benefit from different perspectives on the software, which is augmented by domain specific additional information. This provides them with intuitive access to the structure and behaviour of the software. Various use cases are possible, for example the cooperative analysis of a run time anomaly in a team of domain experts. The domain experts can see the same static structure, augmented with domain specific and detailed information. In the VR environment, they can share their findings and cooperate using their different expertise.
In addition, the static and dynamic properties of the software system are analyzed. Static properties include source code, static call relationships or metrics such as LoC, cyclomatic complexity, etc. Dynamic properties can be grouped into logs, traces, runtime metrics, or configurations that are read in at runtime. The challenge lies in aggregating, analyzing, and correlating this wealth of information. An anomaly and significance detection is developed that automatically detects both structural and runtime anomalies. In addition, a prediction system is set up to make statements about component health. This makes it possible, for example, to predict which components are at risk of failing in the near future. Previously, the log entries were added to the traces, creating a detailed picture of the dynamic call relationships. These dynamic relationships are mapped to the static call graph because they describe calls that do not result from the static analysis (for example, REST calls across several distributed components).
In 2018, the following significant contributions have been made:
- Development of a functional VR visualization prototype for demonstration and research purposes.
- Mapping between dynamic run time data and static structure (required by later analysis and visualization tasks).
- First draft and implementation of the trace anomaly detection by an unsupervised learning procedure. Evaluation and further improvements will follow in the coming months.
In 2019 we achieved the following improvements:
- Extension of the prototype to display dynamic software behaviour.
- Cooperative (remote-)usability of the visualization prototype.
- Interpretation of commit messages for anomaly detection.
- Clustering system calls according to use cases.
Our paper "Towards Collaborative and Dynamic Software Visualization in VR" has been accepted for publication at the International Conference on Computer Graphics Theory and Applications (VISIGRAPP) 2020. It presents the efficiency of our prototype at increasing the software understanding process. In 2020, our paper "A Layered Software City for Dependency Visualization" was accepted at the International Conference on Computer Graphics Theory and Applications (VISIGRAPP) 2021 and later received the "Best Paper Award". We demonstrated that our Layered Layout for Software Cities simplifies the analysis of software architecture and outperforms the standard layout by far. We successfully concluded the research project with a final prototype and the resulting publications.
In 2021, after the end of the official project funding we were asked to submit an extended version of the award paper (" Static And Dynamic Dependency Visualization In A Layered Software City") for review to a journal. Here we present a night view of the city that shows dynamic dependencies as arcs. We thus addressed a central, yet remaining issue: the visualization of dynamic dependencies. In the paper "Trace Visualization within the Software. City Metaphor: A Controlled Experiment on Program Comprehension" at the IEEE Working Conference on Software Visualization (VISSOFT), we displayed dynamic dependencies within the Software City by means of light intensities and were able to show that this representation is more helpful than drawing all dependencies. Also for this paper, we were invited to submit an extended journal article "Trace Visualization within the Software City Metaphor: Controlled Experiments on Program Comprehension" for review. This article demonstrates an extended visualization of dynamic dependencies and color arcs based on HTTP status codes.
In 2022, both journal papers were accepted: "Static And Dynamic Dependency Visualization in a Layered Software City" is published in Springer Nature Computer Science Journal and "Trace Visualization within the Software City Metaphor: Controlled Experiments on Program Comprehension" was accepted for the Information and Software Technology Journal. For the finalization of Holoware, all extensions were combined into one single visualization. For this purpose, different views were applied, allowing the user to switch between them: in the day view, the software architecture can be analyzed in the novel Holoware layered layout, and in the night view, dynamic dependencies are displayed. As part of a master thesis, Holoware was also implemented as an AR visualization, so that it can easily be used as a showcase or in everyday work.
In mid 2023, we finalized the project with the dissertation "Visualizing the statics, dynamics and infrastructure of software using the city metaphor". It summarizes all investigated aspects: (a) the static structure of the system to understand the software architecture, (b) the dynamics of the system to understand the dynamic dependencies (e.g. modern microservice architectures), and (c) the infrastructure of the system to analyze costs and promote the understanding of software operation. We also uncovered another use case: the use of Holoware at trade fairs. The visualization of the software makes it easy to get into conversation with other software developers, as the visualized software can be discussed immediately. To this end, we simplified the setup of the AR and VR visualization so that Holoware can easily be started without a lot of prior technical knowledge. In addition, we improved the contrast of the visualization to make it easier to recognize outlines and arcs, especially in very bright lighting conditions.
Design for Diagnosability
(Third Party Funds Single)Term: 15.05.2013 - 30.09.2018
Funding source: Bayerisches Staatsministerium für Wirtschaft und Medien, Energie und Technologie (StMWIVT) (ab 10/2013)
URL: http://www2.informatik.uni-erlangen.de/research/DfD/Many software systems behave obtrusively during the test phase or even in normal operation. The diagnosis and the therapy of such runtime anomalies is often time consuming and complex, up to being impossible. There are several possible consequences for using the software system: long response times, inexplicable behaviors, and crashes. The longer the consequences remain unresolved, the higher is the accumulated economic damage.
"Design for Diagnosability" is a tool chain targeted towards increasing the diagnosability of software systems. By using the tool chain that consists of modeling languages, components, and tools, runtime anomalies can easily be identified and solved, ideally already while developing the software system. Our cooperation partner QAware GmbH provides a tool called Software EKG that enables developers to explore runtime metrics of software systems by visualizing them as time series.
The research project Design for Diagnosability enhances the eco-system of the existing Software EKG. The Software-Blackbox measures technical and functional runtime values of a software system in a minimally intrusive way. We store the measured values as time series in a newly developed time series database, called Chronix. Chronix is an extremely efficient storage of time series that optimizes disk space as well as response times. Chronix is an open source project (www.chronix.io) and is free to use for everyone.
The newly developed Time-Series-API analyzes these values, e.g., by means of an outlier detection mechanism. The Time-Series-API provides multiple additional building blocks to implement further strategies for identifying runtime anomalies.
The mentioned tools in combination with the existing Software EKG will become the so-called Dynamic Analysis Workbench. This tool enables developers to diagnose, explain, and fix any occurring runtime anomalies both quickly and reliably. It will provide diagnosis plans to localize and identify the root causes of runtime anomalies. The full tool chain aims at increasing the quality of software systems, particularly with respect to the metrics mean-time-to-repair and mean-time-between-defects.
Before we have successfully completed the project in July 2016, we have made the following contributions:
- We have linked Chronix and a framework for distributed data processing so that our anomaly analyses now scale to huge sets of time series data.
- We extended Chronix with additional components. Among them are, for example, a more efficient storage model, some adapters for more time series databases, additional server-side analysis functions, and some new time series types.
- We have published our benchmark for time series databases.
- We have investigated and implemented an approach to link application-level calls, e.g., a login of a user, down to the resulting calls on the OS level.
Although funding expired in 2016, we made further contributions in 2017:
- We presented Chronix at the FAST conference in Santa Clara, CA in February 2017.
- We have equipped Chronix with interfaces to attach time series databases that are used in the industry.
- We have developed an approach that determines the ideal cluster configuration (w.r.t. processing time and costs) for a given analysis (specific function and set of time series).
- We have expanded Spark, a framework for distributed processing of large-scale data, so that it now can make use of GPUs in distributed time series analyses. We presented the results at the Apache Big Data Conference in Miami, Florida, in May 2017.
We continued to make further contributions to the research project in 2018:
- We have published a paper at PROFES 2018 that describes techniques and insights on how runtime data in a large software project can be offered to all project participants at the development stage to improve their collaboration.
- We have maintained the Chronix Open Source project and stabilized it further (updating versions, fixing bugs, etc.).
Effiziente Speicherung von Zeitreihen mit Betriebsdaten aus Software-Systemen zur Analyse von Laufzeitanomalien (Dissertation, 2019)
Making Runtime Data Useful for Incident Diagnosis: An Experience Report
International Conference on Product-Focused Software Process Improvement (PROFES 2018) (Wolfsburg, 28.11.2018 - 30.11.2018)
In: Kuhrmann, Marco; Schneider, Kurt; Pfahl, Dietmar; Amasaki, Sousuke; Ciolkowski, Marcus; Hebig, Regina; Tell, Paolo; Klünder, Jil; Küpper, Steffen (ed.): Proceedings of the International Conference on Product-Focused Software Process Improvement (PROFES 2018), Cham: 2018
Chronix: Long Term Storage and Retrieval Technology for Anomaly Detection in Operational Data
15th USENIX Conference on File and Storage Technologies (FAST 17) (Santa Clara, CA, 27.02.2017 - 02.03.2017)
In: USENIX Association (ed.): Proceedings of the 15th USENIX Conference on File and Storage Technologies (FAST 17) 2017
Open Access: https://www.usenix.org/conference/fast17/technical-sessions/presentation/lautenschlager
, , , :
Fast and efficient operational time series storage: The missing link in dynamic software analysis
Symposium on Software Performance (SSP 2015) (München, 04.11.2015 - 06.11.2015)
In: Softwaretechnik-Trends (Band 35, Nr. 3): Proceedings of the Symposium on Software Performance (SSP 2015) 2015
, , , :
Rahmenwerk zur Ausreißererkennung in Zeitreihen von Software-Laufzeitdaten
Fachtagung Software Engineering & Management (SE 2015) (Dresden, Deutschland, 17.03.2015 - 20.03.2015)
In: Uwe Aßmann, Birgit Demuth, Thorsten Spitta, Georg Püschel, Ronny Kaiser (ed.): Software Engineering & Management (SE 2015), Bonn: 2015
, , , :
Design for Diagnosability: Wie mache ich Software diagnostizierbar?
German Testing Day 2013 (Munich, Germany, 12.11.2013 - 12.11.2013)
In: German Testing Day 2013 2013