TY - GEN A1 - Brand, Thomas A1 - Giese, Holger T1 - Generic adaptive monitoring based on executed architecture runtime model queries and events T2 - IEEE Xplore N2 - Monitoring is a key functionality for automated decision making as it is performed by self-adaptive systems, too. Effective monitoring provides the relevant information on time. This can be achieved with exhaustive monitoring causing a high overhead consumption of economical and ecological resources. In contrast, our generic adaptive monitoring approach supports effectiveness with increased efficiency. Also, it adapts to changes regarding the information demand and the monitored system without additional configuration and software implementation effort. The approach observes the executions of runtime model queries and processes change events to determine the currently required monitoring configuration. In this paper we explicate different possibilities to use the approach and evaluate their characteristics regarding the phenomenon detection time and the monitoring effort. Our approach allows balancing between those two characteristics. This makes it an interesting option for the monitoring function of self-adaptive systems because for them usually very short-lived phenomena are not relevant. Y1 - 2019 SN - 978-1-7281-2731-6 U6 - https://doi.org/10.1109/SASO.2019.00012 SN - 1949-3673 SP - 17 EP - 22 PB - IEEE CY - New York ER - TY - GEN A1 - Ghahremani, Sona A1 - Giese, Holger T1 - Performance evaluation for self-healing systems BT - Current Practice & Open Issues T2 - 2019 IEEE 4th International Workshops on Foundations and Applications of Self* Systems (FAS*W) N2 - Evaluating the performance of self-adaptive systems (SAS) is challenging due to their complexity and interaction with the often highly dynamic environment. In the context of self-healing systems (SHS), employing simulators has been shown to be the most dominant means for performance evaluation. Simulating a SHS also requires realistic fault injection scenarios. We study the state of the practice for evaluating the performance of SHS by means of a systematic literature review. We present the current practice and point out that a more thorough and careful treatment in evaluating the performance of SHS is required. KW - self-healing KW - failure profile KW - evaluation KW - simulator KW - performance Y1 - 2019 SN - 978-1-7281-2406-3 U6 - https://doi.org/10.1109/FAS-W.2019.00039 SP - 116 EP - 119 PB - IEEE CY - New York ER - TY - JOUR A1 - Dyck, Johannes A1 - Giese, Holger A1 - Lambers, Leen T1 - Automatic verification of behavior preservation at the transformation level for relational model transformation JF - Software and systems modeling N2 - The correctness of model transformations is a crucial element for model-driven engineering of high-quality software. In particular, behavior preservation is an important correctness property avoiding the introduction of semantic errors during the model-driven engineering process. Behavior preservation verification techniques show some kind of behavioral equivalence or refinement between source and target model of the transformation. Automatic tool support is available for verifying behavior preservation at the instance level, i.e., for a given source and target model specified by the model transformation. However, until now there is no sound and automatic verification approach available at the transformation level, i.e., for all source and target models. In this article, we extend our results presented in earlier work (Giese and Lambers, in: Ehrig et al (eds) Graph transformations, Springer, Berlin, 2012) and outline a new transformation-level approach for the sound and automatic verification of behavior preservation captured by bisimulation resp.simulation for outplace model transformations specified by triple graph grammars and semantic definitions given by graph transformation rules. In particular, we first show how behavior preservation can be modeled in a symbolic manner at the transformation level and then describe that transformation-level verification of behavior preservation can be reduced to invariant checking of suitable conditions for graph transformations. We demonstrate that the resulting checking problem can be addressed by our own invariant checker for an example of a transformation between sequence charts and communicating automata. KW - Relational model transformation KW - Formal verification of behavior preservation KW - Behavioral equivalence and refinement KW - Bisimulation and simulation KW - Graph transformation KW - Triple graph grammars KW - Invariant checking Y1 - 2018 U6 - https://doi.org/10.1007/s10270-018-00706-9 SN - 1619-1366 SN - 1619-1374 VL - 18 IS - 5 SP - 2937 EP - 2972 PB - Springer CY - Heidelberg ER - TY - BOOK A1 - Giese, Holger A1 - Maximova, Maria A1 - Sakizloglou, Lucas A1 - Schneider, Sven T1 - Metric temporal graph logic over typed attributed graphs BT - extended version N2 - Graph repair, restoring consistency of a graph, plays a prominent role in several areas of computer science and beyond: For example, in model-driven engineering, the abstract syntax of models is usually encoded using graphs. Flexible edit operations temporarily create inconsistent graphs not representing a valid model, thus requiring graph repair. Similarly, in graph databases—managing the storage and manipulation of graph data—updates may cause that a given database does not satisfy some integrity constraints, requiring also graph repair. We present a logic-based incremental approach to graph repair, generating a sound and complete (upon termination) overview of least-changing repairs. In our context, we formalize consistency by so-called graph conditions being equivalent to first-order logic on graphs. We present two kind of repair algorithms: State-based repair restores consistency independent of the graph update history, whereas deltabased (or incremental) repair takes this history explicitly into account. Technically, our algorithms rely on an existing model generation algorithm for graph conditions implemented in AutoGraph. Moreover, the delta-based approach uses the new concept of satisfaction (ST) trees for encoding if and how a graph satisfies a graph condition. We then demonstrate how to manipulate these STs incrementally with respect to a graph update. N2 - Verschiedene Arten typisierter attributierter Graphen können verwendet werden, um Systemzustände aus einem breiten Bereich von Domänen darzustellen. Für dynamische Systeme können etablierte Formalismen wie die Graphtransformation ein formales Modell für die Definition von Zustandssequenzen liefern. Wir betrachten den Fall, in dem zwischen Zustandsänderungen Zeit vergehen kann, und führen eine Logik ein, die als Metric Temporal Graph Logic (MTGL) bezeichnet wird, um über solche zeitgesteuerten Graphsequenzen zu urteilen. Mit dieser Logik drücken wir Eigenschaften der Struktur und der Attribute von Zuständen sowie des Auftretens von Zuständen über die Zeit aus, die durch ihre innere Struktur miteinander verbunden sind, was bisher keine formale Logik über Graphen präzise bewerkstelligt. Erstens, basierend auf zeitgesteuerten Graphsequenzen als Modelle für die Systemevolution, definieren wir MTGL, indem wir den zeitlichen Operator bis zu einer gewissen Zeitgrenze in die etablierte Logik von (verschachtelten) Graphbedingungen integrieren. Zweitens skizzieren wir, wie eine endliche zeitgesteuerte Diagrammsequenz als einzelnes Diagramm dargestellt werden kann, das alle zeitlichen Änderungen enthält (als Diagramm mit Verlauf bezeichnet), wie die Erfüllung von MTGL-Bedingungen für ein solches Diagramm definiert werden kann, und zeigen, dass beide Darstellungen dieselben MTGL-Bedingungen erfüllen. Drittens zeigen wir, wie MTGL-Bedingungen auf (verschachtelte) Diagrammbedingungen reduziert werden können, und zeigen anhand dieser Reduzierung, dass beide zugrunde liegenden Logiken gleichermaßen aussagekräftig sind. Schließlich stellen wir eine Erweiterung des Tools AutoGraph vor, mit der die Erfüllung der MTGL-Bedingungen für zeitgesteuerte Diagrammsequenzen überprüft werden kann, indem die Erfüllung der (verschachtelten) Diagrammbedingungen überprüft wird, die unter Verwendung der vorgeschlagenen Reduzierung für das Diagramm mit dem Verlauf entsprechend dem zeitgesteuerten Diagramm erhalten wurden. T3 - Technische Berichte des Hasso-Plattner-Instituts für Digital Engineering an der Universität Potsdam - 127 KW - typisierte attributierte Graphen KW - metrisch temporale Graph Logic KW - Spezifikation von gezeiteten Graph Transformationen KW - typed attributed graphs KW - metric termporal graph logic KW - specification of timed graph transformations Y1 - 2019 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:517-opus4-427522 SN - 978-3-86956-463-0 SN - 1613-5652 SN - 2191-1665 IS - 127 PB - Universitätsverlag Potsdam CY - Potsdam ER -