The ITEC team participated in the HiPeac 2020 International Workshop on Exascale programing models for extreme data with a presentation with title “Monitoring data collection and mining for Exascale systems”. The ITEC team also attended the collocated ASPIDE meeting and actively participated in the decision of the next research activities in the project.
Title of the talk: Mobility-Aware Scheduling of Extreme Data Workflows across the Computing Continuum
Abstract: The appearance of the Fog/Edge computing paradigm, as an emanation of the computing continuum closer to the edge of the network, unravels important opportunities for execution of complex business and scientific workflows near the data sources. The main characteristics of these workflows are (i) their distributed nature, (ii) the vast amount of data (in the order of petabytes) they generate and (iii) the strict latency requirements. Current workflow management approaches rely exclusively on the Cloud Data Centers, which due to their geographical distance in relation to the data sources, could negatively influence the latency and cause violation of workflow requirements. It is therefore essential to research novel concepts for partial offloading of complex workflows closer to where the data is generated, thus reducing the communication latency and the need for frequent data transfers.
In this talk we will explore the potential of the computing continuum for scheduling and partial offloading of complex workflows with strict response time requirements and expose the resource provisioning challenges related to the heterogeneity and mobility of the Fog/Edge environment. Consequently, we will discuss a novel mobility-aware Pareto-based approach for task offloading across the continuum, which considers three optimization objectives, namely response time, reliability, and financial cost. Besides, the approach introduces a Markov model to perform a single-step predictive analysis on the mobility of the Fog/Edge devices, thus constraining the task offloading optimization problem to devices that do not frequently move (roam) within the computing continuum. As a conclusion to the talk, we will discuss the efficiency of the presented approach, based on both a simulated and a real-world testbed environment tailored for a set real-world biomedical, meteorological and astronomy workflows.
The paper “Making simulation results reproducible—Survey, guidelines, and examples based on Gradle and Docker” has been accepted and published in PeerJ Computer Science.
Authors: Wilfried Elmenreich, Philipp Moll, Sebastian Theuermann, Mathias Lux (Alpen-Adria-Universität Klagenfurt)
Abstract: This article addresses two research questions related to reproducibility within the context of research related to computer science. First, a survey on reproducibility addressed to researchers in the academic and private sectors is described and evaluated. The survey indicates a strong need for open and easily accessible results, in particular, reproducing an experiment should not require too much effort. The results of the survey are then used to formulate guidelines for making research results reproducible. In addition, this article explores four approaches based on software tools that could bring forward reproducibility in research results. After a general analysis of tools, three examples are further investigated based on actual research projects which are used to evaluate previously introduced tools. Results indicate that the evaluated tools contribute well to making simulation results reproducible but due to conflicting requirements, none of the presented solutions fulfills all intended goals perfectly.
The full paper can be found on: https://peerj.com/articles/cs-240/
Authors: Hadi Amirpour, Ekrem Çetinkaya (Alpen-Adria-Universität Klagenfurt), Christian Timmerer (Alpen-Adria-Universität Klagenfurt, Bitmovin), and Mohammad Ghanbari (University of Tehran, University of Essex)
Abstract: Adaptive HTTP streaming is the preferred method to deliver multimedia content on the internet. It provides multiple representations of the same content in different qualities (i.e., bit-rates and resolutions) and allows the client to request segments from the available representations in a dynamic, adaptive way depending on its context. The growing number of representations in adaptive HTTP streaming makes encoding of one video segment at different representations a challenging task in terms of encoding time-complexity. In this paper, information of both highest and lowest quality representations are used to limit Rate Distortion Optimization (RDO) for each Coding Unit Tree (CTU) in High Efficiency Video Coding. Our proposed method first encodes the highest quality representation and consequently uses it to encode the lowest quality representation. In particular, the block structure and the selected reference frame of both highest and lowest quality representations are then used to predict and shorten the RDO process of each CTU for intermediate quality representations. Our proposed method introduces a delay of two CTUs thanks to employing parallel processing techniques. Experimental results show a significant reduction in time-complexity over the reference software (38%) and state-of-the-art (10%) is achieved while quality degradation is negligible.
Keywords: HTTP adaptive streaming, Multi-rate encoding, HEVC, Fast block partitioning
The PhD defense is in progress. @ilyushkin of @LargeResearch and @EEMCS_TUD discusses his work with @RaduProdanAAU. Science at work. pic.twitter.com/ubSZoXRYCR
— Alexandru Iosup (@AIosup) December 19, 2019
The paper has been accepted (through double-blind peer review) as a regular paper of the Euromicro PDP’2020 conference to be held in Vasteras, Sweden on 11-13 March, 2020.
Title: M3AT: Monitoring Agents Assignment Model for Data-Intensive Applications
Authors: Vladislav Kashansky, Dragi Kimovski, Radu Prodan, Prateek Agrawal, Fabrizio Marozzo, Iuhasz Gabriel, Marek Justyna and Javier Garcia-Blas
Abstract: Nowadays, massive amounts of data are acquired, transferred, and analyzed nearly in real-time by utilizing a large number of computing and storage elements interconnected through high-speed communication networks. However, one issue that still requires research effort is to enable efficient monitoring of applications and infrastructures of such complex systems. In this paper, we introduce a Integer Linear Programming (ILP) model called M3AT for optimised assignment of monitoring agents and aggregators on large-scale computing systems. We identified a set of requirements from three representative data-intensive applications and exploited them to define the model’s input parameters. We evaluated the scalability of M3AT using the Constraint Integer Programing (SCIP) solver with default configuration based on synthetic data sets. Preliminary results show that the model provides optimal assignments for systems composed of up to 200 monitoring agents while keeping the number of aggregators constant and demonstrates variable sensitivity with respect to the scale of monitoring data aggregators and limitation policies imposed.
Keywords: Monitoring systems, high performance computing, aggregation, systems control, data-intensive systems, generalized assignment problem, SCIP optimization suite.
Acknowledgement: This work has received funding from the EC-funded project H2020 FETHPC ASPIDE (Agreement #801091)
Narges Mehran presented the paper “MAPO: A Multi-Objective Model for IoT Application Placement in a Fog Environment” at the 9th International Conference on the Internet of Things, IoT 2019 in Bilbao, Spain (October 22-25, 2019).
Authors: Narges Mehran, Dragi Kimovski, Radu Prodan (Alpen-Adria Universität Klagenfurt).
Abstract: The emergence of the Fog computing paradigm that leverages in-network virtualized resources raises important challenges in terms of resource and IoT application management in a heterogeneous environment with limited computing resources. In this work, we propose a novel Pareto-based approach for application placement close to the data sources called Multi-objective IoT Application Placement in fOg (MAPO). MAPO models applications based on a finite state machine using three conflicting optimization objectives, completion time, energy consumption, and economic cost, and considering both the computation and communication aspects. In contrast to existing solutions that optimize a single objective, MAPO enables multi-objective energy and cost-aware application placement. To evaluate the quality of the MAPO placements, we created both simulated and real-world testbeds tailored for a set of medical IoT application case studies. Compared to the state-of-the-art approaches, MAPO reduces the economic cost by 28%, while decreasing the energy requirements by 29-64% on average, and improves the completion time by a factor of six.
Track: IoT Edge and Cloud @IoT’19
Acknowledgement: Austrian Research Promotion Agency (FFG), project 848448, Tiroler Cloud, funded this work.
Prof. Radu Prodan participated at the PhD examination of Dr. Yang Hu and Dr. Huan Zhou at the University of Amsterdam, Netherlands.
With: Prof. Peter van Emde Boas, Prof. Cees de Laat, Prof. Henri E. Bal, Dr. Paola Grosso, Dr. Zhiming Zhao, Prof. Rob van Nieuwpoort, Prof. Pieter Adriaans, Dr. Adam S.Z. Belloum
Sabrina Kletz presented the paper “Learning the Representation of Instrument Images in Laparoscopy Video” at the MIAR Workshop @ MICCAI 2019 in Shenzhen, China.
Authors: Sabrina Kletz, Klaus Schoeffmann, Heinrich Husslein
Abstract: Automatic recognition of instruments in laparoscopy videos poses many challenges that need to be addressed like identifying multiple instruments appearing in various representations and in different lighting conditions which in turn may be occluded by other instruments, tissue, blood or smoke. Considering these challenges it may be beneficial for recognition approaches that instrument frames are first detected in a sequence of video frames for further investigating only these frames. This pre-recognition step is also relevant for many other classification tasks in laparoscopy videos such as action recognition or adverse event analysis. In this work, we address the task of binary classification to recognize video frames as either instrument or non-instrument images. We examine convolutional neural network models to learn the representation of instrument frames in videos and take a closer look at learned activation patterns. For this task, GoogLeNet together with batch normalization is trained and validated using a publicly available dataset for instrument count classifications. We compare transfer learning with learning from scratch and evaluate on datasets from cholecystectomy and gynecology. The evaluation shows that fine-tuning a pre-trained model on the instrument and non-instrument images is much faster and more stable in learning than training a model from scratch.
Conference: 2019 International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), October 13–17, 2018, Shenzhen, China
Track: Medical Imaging and Augmented Reality (MIAR) Workshop @MICCAI