Kálmán Képes

Combining Declarative and Imperative Cloud Application Provisioning Based on TOSCA

The automation of application provisioning is one of the most important issues in Cloud Computing. The Topology and Orchestration Specification for Cloud Applications (TOSCA) supports automating provisioning by two different flavors: (i) declarative processing is based on interpreting application topology models by a runtime that infers provisioning logic whereas (ii) imperative processing employs provisioning plans that explicitly describe the provisioning tasks to be executed. Both flavors come with benefits and drawbacks. This paper presents a means to combine both flavors to resolve drawbacks and to profit from benefits of both worlds: we propose a standards-based approach to generate provisioning plans based on TOSCA topology models. These provisioning plans are workflows that can be executed fully automatically and may be customized by application developers after generation. We prove the technical feasibility of the approach by an end-to-end open source toolchain and evaluate its extensibility, performance, and complexity.

OpenTOSCA for IoT: Automating the Deployment of IoT Applications based on the Mosquitto Message Broker

Automating the deployment of IoT applications is a complex challenge, especially if multiple heterogeneous sensors, actuators, and business components have to be integrated. This demonstration paper presents a generic, standards-based system that is able to fully automatically deploy IoT applications based on the TOSCA standard, the standardized MQTT messaging protocol, the Mosquitto message broker, and the runtime environment OpenTOSCA. We describe a demonstration scenario and explain in detail how this scenario can be deployed fully automatically using the mentioned technologies.

Automating the Provisioning and Configuration of Devices in the Internet of Things

The Internet of Things benefits from an increasing number of interconnected technical devices. This has led to the existence of so-called smart environments, which encompass one or more devices sensing, acting, and automatically performing different tasks to enable their self-organization. Smart environments are divided into two parts: the physical environment and its digital representation, oftentimes referred to as digital twin. However, the automated binding and monitoring of devices of smart environments are still major issues. In this article we present a method and system architecture to cope with these challenges by enabling (i) easy modeling of sensors, actuators, devices, and their attributes, (ii) dynamic device binding based on their type, (iii) the access to devices using different paradigms, and (iv) the monitoring of smart environments in regard to failures or changes. We furthermore provide a prototypical implementation of the introduced approach.

A situation-aware workflow modelling extension

The automation of business processes is of vital importance for organizations to speed up their business and to lower costs. Due to emerging technologies in the field of Internet of Things, changing situations can be recognized automatically, which provides the basis for an automated adaptation of process executions in order to react to changing circumstances. Although approaches exist that enable creating self-adapting workflows, a systematic modelling approach that supports the specification of situational dependencies directly in workflow models is missing. In this paper, we tackle this issue by presenting a modelling extension called SitME that defines (i) an extensible Situation Event type, (ii) the concept of Situational Scopes, and (iii) a visual notation. As the introduced extension is language-independent, we apply the approach to BPEL to validate its practical feasibility.

OpenTOSCA for the 4th Industrial Revolution: Automating the Provisioning of Analytics Tools based on Apache Flink

The 4th industrial revolution entails new levels of data driven value chain organization and management. In industrial environments, the optimization of whole production lines based on machine learning algorithms allow to generate huge business value. Still, one of the open challenges is how to process the collected data as close to the data sources as possible. To fill this gap, this paper presents an OpenTOSCA-based toolchain that is capable of automatically provisioning Apache Flink as a holistic analytics environment altogether with specialized machine learning algorithms. This stack can be deployed as close to the production line as possible to enable data driven optimization. Further, we demonstrate how the analytics stack can be modeled based on TOSCA to be automatically provisioned considering specific mock services to simulate machine metering in the development phase of the algorithms.

Situation-Aware Execution and Dynamic Adaptation of Traditional Workflow Models

The continuous growth of the Internet of Things together with the complexity of modern information systems results in several challenges for modeling, provisioning, executing, and maintaining systems that are capable of adapting themselves to changing situations in dynamic environments. The properties of the workflow technology, such as its recovery features, makes this technology suitable to be leveraged in such environments. However, the realization of situation-aware mechanisms that dynamically adapt process executions to changing situations is not trivial and error prone, since workflow modelers cannot reflect all possibly occurring situations in complex environments in their workflow models. In this paper, we present a method and concepts to enable modelers to create traditional, situation-independent workflow models that are automatically transformed into situation-aware workflow models that cope with dynamic contextual situations. Our work builds upon the usage of workflow fragments, which are dynamically selected during runtime to cope with prevailing situations retrieved from low-level context sensor data. We validate the practical feasibility of our work by a prototypical implementation of a Situation-aware Workflow Management System (SaWMS) that supports the presented concepts.

Hybrid TOSCA Provisioning Plans: Integrating Declarative and Imperative Cloud Application Provisioning Technologies

The efficient provisioning of complex applications is one of the most challenging issues in Cloud Computing. Therefore, various provisioning and configuration management technologies have been developed that can be categorized as follows: imperative approaches enable a precise specification of the low-level tasks to be executed whereas declarative approaches focus on describing the desired goals and constraints. Since complex applications employ a plethora of heterogeneous components that must be wired and configured, typically multiple of these technologies have to be integrated to automate the entire provisioning process. In a former work, we presented a workflow modelling concept that enables the seamless integration of imperative and declarative technologies. This paper is an extension of that work to integrate the modelling concept with the Cloud standard TOSCA. In particular, we show how Hybrid Provisioning Plans can be created that retrieve all required information about the desired provisioning directly from the corresponding TOSCA model. We validate the practical feasibility of the concept by extending the OpenTOSCA runtime environment and the workflow language BPEL.

OpenTOSCA Injector: Vertical and Horizontal Topology Model Injection

The automation of application deployments is supported by various technologies. The TOSCA standard facilitates to describe application deployments in a portable manner by modeling application structures as topology models. The final structure often depends on the target environment and is, therefore, not always known at modeling time. However, a manual adaptation is error-prone and time-consuming. In this paper, we demonstrate the OpenTOSCA Injector for an automated completion of topology models: the extended TOSCA runtime OpenTOSCA for an automated injection and deployment is presented.