Kropf, Peter

2020-6-1, Kropf, Peter, Lapin, Andrei, Carretero, Jesus, Caíno-Lores, Silvina

The increasing amounts of data related to the execution of scientific workflows has raised awareness of their shift towards parallel data-intensive problems. In this paper, we deliver our experience combining the traditional high-performance computing and grid-based approaches with Big Data analytics paradigms, in the context of scientific ensemble workflows. Our goal was to assess and discuss the suitability of such data-oriented mechanisms for production-ready workflows, especially in terms of scalability. We focused on two key elements in the Big Data ecosystem: the data-centric programming model, and the underlying infrastructure that integrates storage and computation in each node. We experimented with a representative MPI-based iterative workflow from the hydrology domain, EnKFHGS, which we re-implemented using the Spark data analysis framework. We conducted experiments on a local cluster, a private cloud running OpenNebula, and the Amazon Elastic Compute Cloud (AmazonEC2). The results we obtained were analysed to synthesize the lessons we learned from this experience, while discussing promising directions for further research.

2014, Felber, Pascal, Kropf, Peter, Schiller, Eryk, Serbu, Sabina

2010-1-22, Casera, Steve, Kropf, Peter

Mechanisms and tools able to facilitate collaboration are needed when remotely team members wish to jointly analyze data to create and acquire new knowledge. This paper discusses the main challenges involved in remote collaboration systems used in scientific visualization, and presents our solutions and assessments. To evaluate the usability and effectiveness of the collaborative tools comprising this system, we conduct an experiment on several groups of users. More particularly, we compare various forms of audio and textual communication tools as well as graphical indicators used to exchange information. Our analyses reveal how important it is to choose the means of conversation and, surprisingly, that graphical indicators have little influence. Furthermore, these experiments demonstrate that the system’s usability can be significantly improved through providing appropriate grounding and awareness support. This results in increased satisfaction, and in turn a potential increase the quality of collaborative work.

2008, Maret, Timothée, Kummer Raphaël, Kropf, Peter, Wagen, Jean-Frédéric

Research on Wireless Sensor Networks (WSNs) has developed highly optimized software environments fitting the limited hardware resource constraints of Motes. Unfortunately, these environments suffer from relatively complex programming models. Nowadays well known languages such as Java and optimized JVMs become available and simplify the application development for the Motes. Thus, we developed the Freemote Emulator which is a Java based emulator providing a lightweight emulation tool for emerging Java based Motes. It runs experiments in real time mixing real and emulated nodes. Its layered architecture and a set of predefined code templates allow developers to quickly produce runnable code for real and emulated nodes as well as predefined scenarios to help the newcomers to introduce into the system and WSNs. Our emulator provides as well a useful visualization tool based on a parametrizable slow down feature that helps to understand complex WSN behaviours and to debug tricky implementation problems. Finally, a single emulation can run on several computers, thus allowing programmers to conduct experiments with a pretty large number of emulated and real nodes.

2017-7-1, Kropf, Peter, Kurtz, Wolfgang, Lapin, Andrei, Tang, Qi, Schilling, Oliver, Schiller, Eryk, Braun, Torsten, Hunkeler, Daniel, Vereecken, Harry, Sudicky, Edward, Franssen, Harrie-Jan Hendricks, Brunner, Philip

Online data acquisition, data assimilation and integrated hydrological modelling have become more and more important in hydrological science. In this study, we explore cloud computing for integrating field data acquisition and stochastic, physically-based hydrological modelling in a data assimilation and optimisation framework as a service to water resources management. For this purpose, we developed an ensemble Kalman filter-based data assimilation system for the fully-coupled, physically-based hydrological model HydroGeoSphere, which is able to run in a cloud computing environment. A synthetic data assimilation experiment based on the widely used tilted V-catchment problem showed that the computational overhead for the application of the data assimilation platform in a cloud computing environment is minimal, which makes it well-suited for practical water management problems. Advantages of the cloud-based implementation comprise the independence from computational infrastructure and the straightforward integration of cloud-based observation databases with the modelling and data assimilation platform.

2013, Felber, Pascal, Kropf, Peter, Leonini, Lorenzo, Luu, Toan, Rajman, Martin, Rivière, Etienne, Schiavoni, Valerio, Valerio, José

2008, Tutsch, Dietmar, Babin, Gilbert, Kropf, Peter

Self-similar or multifractal behavior has been observed for LAN and Internet WAN (backbone) traffic. Investigations about this kind of behavior for application level protocols are rarely found because sessions or even applications are usually too short to be characterized in this direction. Only Telnet and FTP were examined so far. This paper analyzes the traffic shape of Peer-to-Peer (P2P) networks using the Gnutella protocol. Data was collected using a modified LimeWire servent. Self-similarity was estimated using a variance-time plot. The results show that Gnutella messages exhibit a self-similar shape, regardless of the message type.

2015, Kropf, Peter

Studies of complex physical and engineering systems, represented by multi-scale and multi-physics computer simulations have an increasing demand for computing power, especially when the simulations of realistic problems are considered. This demand is driven by the increasing size and complexity of the studied systems or the time constraints. Ultrascale computing systems offer a possible solution to this problem. Future ultrascale systems will be large-scale complex computing systems combining technologies from high performance computing, distributed systems, big data, and cloud computing. Thus, the challenge of developing and programming complex algorithms on these systems is twofold. Firstly, the complex algorithms have to be either developed from scratch, or redesigned in order to yield high performance, while retaining correct functional behaviour. Secondly, ultrascale computing systems impose a number of non-functional cross-cutting concerns, such as fault tolerance or energy consumption, which can significantly impact the deployment of applications on large complex systems. This article discusses the state-of-the-art of programming for current and future large scale systems with an emphasis on complex applications. We derive a number of programming and execution support requirements by studying several computing applications that the authors are currently developing and discuss their potential and necessary upgrades for ultrascale execution.

2011, Serbu, Sabina, Felber, Pascal, Kropf, Peter

2008, Tutsch, Dietmar, Babin, Gilbert, Kropf, Peter

Characterizing traffic behavior helps to optimize the network architecture for improved performance. Using a modified LimeWire servent ( for both the server and client) and a variance-time plot for traffic characterization, the authors analyze the Gnutella protocol's traffic shape and find that the messages exhibit a self-similar shape. This result shows network designers that they need to consider the self-similar traffic shape in their set-up - for instance, by introducing appropriate buffer sizes.