Christian Baun’s research while affiliated with Frankfurt University of Applied Sciences and other places

... At the Frankfurt University of Applied Sciences, the SKILL/VL project (Strategic Competence Platform -Innovative Learning and Teaching -Virtualization of Distributed Environments for Teaching) [8] has implemented a digital offer for teaching. The platform enables lecturers, students, and staff to create virtual landscapes to integrate them into their teaching. ...

Reference:

A novel concept for the migration of workloads using virtual machines that can mitigate limitations of traditional service deployment concepts

... Clusters of single board computers have less purchase cost and are easy to operate, and therefore they are a good option for education and research projects with less funding [2]. The Faculty of Computer Science and Engineering of the Frankfurt University of Applied Sciences possesses a cluster of 128 nodes of Raspberry Pi 3 single board computers with 512 physical CPU cores [6]. This cluster serves as a testbed for scientific research and education, and gives students the opportunity to experiment with different parallel computing models. ...

Reference:

A Lightweight Network-Controlled Power Strip for Low-Cost Cluster Systems

... Ein Risiko von Cloud-Services stellt jedoch der mögliche Verlust von Unternehmensgeheimnissen und der Missbrauch von Endkundendaten dar [HC15]. [BBKS15]. Der in der vorliegenden Arbeit verfolgte Ansatz zur effizienten Parallelisierung kann somit als eine organisationsinterne Cloud-Umgebung durch Nutzung der vorhandenen dezentralen Ressourcen angesehen werden. ...

Reference:

Effizienter Einsatz von Optimierungsmethoden in der Produktentwicklung durch dynamische Parallelisierung

... Weiss 2007]. Die Betriebsmodi können hierbei "Public Cloud" (Betrieb bei einem externen Anbieter), "Private Cloud" (Betrieb der Cloud im eigenen Unternehmen) oder "Hybrid Cloud", bei der die Grenzen zwischen den vorgenannten Modi verschwimmen, sein [Baun et al. 2011]. ...

Reference:

KonfIT-SSC: Entwicklung einer Konfigurationsumgebung für IT-Service-Supply-Chains kleiner Unternehmen der Kreativwirtschaft

... Those characteristics show the hardware limitations of the system but yet it is capable of running different Linux distributions. The hardware resources are also sufficient to use them for running object-based cloud storage services [4][5] and for the investigation of the scalability of parallel computation tasks [3]. With a cluster of low-cost nodes like Raspberry Pi 3 computers, the crash of nodes during operation is a common event. ...

Reference:

A Lightweight Network-Controlled Power Strip for Low-Cost Cluster Systems

... Eine Realisierung unabhängiger Teilsysteme für die Identifikation und Aktualisierung von Modellen bzw. für die konkrete Überwachung von Systemen und Prozessen eröffnet folgende Vorteile [P6-P9, P14, P15][S2, S4, S7, S15, S28] [24,37,266,268,306]: ...

Reference:

Modellbasierte Zustands- und Prozessüberwachung auf Basis sozio-cyber-physischer Systeme

... The emergence of the Pi2B in 2015, which boasted a quad-core processor, provided a significant performance boost over the Pi1B, as evidenced by the studies conducted by Christian Baun 9 and Cloutier et al. 10 . Baun compared 8-node Pi1B and Pi2B clusters, finding that Pi2B demonstrated approximately four times higher performance with 80% memory usage. ...

Reference:

Performance evaluation and analysis of scalable Raspberry Pi 4 Model B clusters

... RPi cluster with open-source libraries such as Hadoop and OpenCV could play a pivotal role in image processing e.g. feature points extraction system using Speeded Up Robust Features (SURF) algorithm [27], face recognition [28], image stacking [29], edge detection [30], [31], image analysis [32], ray tracing [33], image conversion [34], image recognition [35], Fourier transform [36] and image classification [37]. Moreover, a RPi cluster can extend its capability in image processing by adding a GPU [38]. ...

Reference:

Distributed Multi-Ant Colony System Algorithm using Raspberry Pi Cluster for Travelling Salesman Problem

... For example, a level game model was presented that looked at how user task execution influenced resource selfishness in multiple administrative domains, but it ignored the connection between resource suppliers. The literature has also proposed a distributed actively noncooperative abandon model [15], which was able to reduce task execution time due to priority resource allocation. However, it is not clear if the Nash equilibrium solution is the only possible option in a pricing approach based on game theory that aims to solve the Nash bargaining dilemma. ...

Reference:

Complexity and Monitoring of Economic Operations Using a Game-Theoretic Model for Cloud Computing

... Because of advantages offered by the hybrid cloud [1], [2], [3], many organizations are adopting a cloud deployment model that allows for easy and rapid scalability, secures control over sensitive business data, and also enables privacy in processing such sensitive data 1 . The compound of the public and private cloud, however, introduces some challenges [2], [3], [4], [5], [6], [7]. Among them, we may stress the differences in the level of access and control of their underlying infrastructure: there is total control in an on-premises private cloud whilst the level of control over the provisioned public cloud resource is relatively limited depending on the service model. ...

Reference:

Modeling and Simulating a Process Mining-Influenced Load- Balancer for the Hybrid Cloud