Hasso-Plattner-InstitutSDG am HPI
Hasso-Plattner-InstitutDSG am HPI
Login
 

Current Projects

The current period is Fall 2021 and started on November 23, 2021. The end is the next HPI Future SOC Lab Day on April 26, 2022.
28 research projects are using the Lab's IT infrastructure. If you would like to receive more information about one or more projects, please contact us.

Germany

Machine Learning to scale telemedical interventions for cardiovascular diseases

Abstract

Cardiovascular diseases are the leading cause of death globally. Telemedicine interventions were shown to reduce the percentage of days lost due to unplanned cardiovascular hospital admissions and all-cause mortality. This project will train an AI-System that could help to scale such telemedical interventions by preprocessing and prioritizing.

 

Researchers

Principle Investigator: Prof. Dr. Andreas Polze || Contact Author: Jossekin Beilharz || Hasso Plattner Institute, Germany

Factorization of Export Ciphers (student project Kryptanalytische Angriffe auf Internet Protokolle)

Abstract

Many Internet protocols make use of legacy components such as outdated cryptography. In this project, students build a proof-of-concept attack how an adversary could attack this and decode TLS traffic in real-time, based on the now weak RSA Export ciphers. For this demo, it is necessary to factorize these old RSA primes.

 

Researchers

Principle Investigator: Christian Dörr || Contact Author: Hadrian Burkhardt || Hasso Plattner Institute, Germany

Mining Flood Insurance Big Data to Reveal the Determinants of Humans' Flood Resilience

Abstract

Human behavior has shown to have a significant impact on future flood risk. The US National Flood Insurance Program has recently released data on flood insurance policies. This study contributes a data-driven approach to identify the main determinants and dynamics of flood insurance purchase throughout different states and social backgrounds.

 

Researchers

Principle Investigator: Prof. Dr. Andrea Cominola || Contact Author: Nadja Veigel || Technical University of Berlin, Germany

PRESLEY - Page Replication for Scale-Up Systems

Abstract

We want to evaluate how page replication as another management mechanism is worth the effort. The experiments are based on a patched linux kernel running on multi-socket NUMA systems. The outcome should be to identify possible workloads and data structures that likely benefit from the replication of pages.

 

Researchers

Principle Investigator: Prof. Dr. Andreas Polze || Contact Author: Felix Eberhardt || Hasso Plattner Institute, Germany

Integrating Hardware Accelerators in Virtualized Environments

Abstract

In this project, we study mechanisms for integrating hardware accelerators in virtual machines and cloud infrastructures. Exemplary workloads include In-Memory Databases, scientific computation and multimedia applications. This project is a continuation of preceding projects conducted since the Spring Period of 2018.

 

Researchers

Principle Investigator: Prof. Dr. Andreas Polze || Contact Author: Max Plauth || Hasso Plattner Institute, Germany

An Energy-Aware Runtime System for Heterogeneous Clusters

Abstract

We are planning to extend our work on PINPOINT (published at the Runtime and Operating Systems for Supercomputers workshop at Super Computing 2020) for energy-efficient and economic processing on heterogeneous compute infrastructures in the HPI Future SOC Lab in cooperation with the OSM Group (HPI).

 

Researchers

Principle Investigator: Prof. Dr.-Ing. habil Wolfgang Schröder-Preikschat || Contact Author: Dr.-Ing. Timo Hönig || Friedrich-Alexander-Universität Erlangen-Nürnberg, Germany

Quality Engineering for Microservices and DevOps

Abstract

Microservices and DevOps are gaining considerable attraction. We would use the requested HPI Future SOC Lab resources to investigate the experimental evaluation of our activities on "DevOps-oriented Load and Resilience Testing for Microservices" and "Automated Cross-Component Issue Classification for Microservices".

 

Researchers

Principle Investigator: Dr.-Ing. Andre van Hoorn || Contact Author: Dr.-Ing. Andre van Hoorn || University of Stuttgart, Germany

Query-Driven Partial Database Replication

Abstract

Partial database replication is a query-driven approach to minimize the overall memory consumption of a replication cluster while still enabling a balanced load distribution among nodes. In this Future SOC Lab project, we want to deploy partial data allocations for large database clusters and evaluate their end-to-end performance.

 

Researchers

Principle Investigator: Prof. Dr. h.c. Hasso Plattner || Contact Author: Stefan Halfpap || Hasso Plattner Institute, Germany

Benchmarking EAs and GNNs on Networks

Abstract

We continue to investigate how algorithmic graph-based techniques might help to answer questions related to epidemics. We benchmark heuristics, some of which are based on Evolutionary Algorithms (EAs) and Reinforcement Learning with Graph Neural Networks (GNNs). We test different configurations on small and large synthetic and real-world graphs.

 

Researchers

Principle Investigator: Prof. Dr. Tobias Friedrich || Contact Author: Karen Seidel || Hasso Plattner Institute, Germany

Blood Pressure Estimation using Photoplethysmography

Abstract

Treating and preventing high blood pressure requires continuous monitoring. Leveraging photoplethysmography (PPG) sensors contained on wearables, machine learning algorithms can be used to estimate systolic as well as diastolic blood pressure values continuously and unobtrusively.

 

Researchers

Principle Investigator: Prof. Dr. med. Erwin Böttinger || Contact Author: Florian Hermes || Hasso Plattner Institute, Germany

Application of Elastic Reverse Time Migration to Ultrasonic Echo Data in Civil Engineering

Abstract

To improve ultrasonic imaging of concrete structures, we transferred the seismic migration technique Reverse Time Migration (RTM) to non-destructive testing. Applying computationally intensive elastic RTM algorithms to ultrasonic data significantly improves imaging of complex features in concrete and needs to be further evaluated and optimized.

Researchers

Principle Investigator: Priv.-Doz. Dr. rer. nat. Ernst Niederleithinger || Contact Author: M. Sc. Maria Grohmann || Federal Institute of Materials Research and Testing, BAM, Germany

Flexible Edge Orchestration Framework

Abstract

We design and develop EdgeIO, a hybrid resource orchestration and application deployment platform that seamlessly combines cloud and edge environments. The primary objective of EdgeIO is to impart flexibility -  in design, operation, moderation and modification, all the while fully exploiting the capabilities of edge & cloud computing.

Researchers

Principle Investigator: Prof. Dr.-Ing. Jörg Ott || Contact Author: Dr. Nitinder Mohan || Technische Universität München, Germany

Europe

CitySensing - Big IoT and mobility data processing and analytics in Smart Cities

Abstract

The project objectives are focused on research and development of methods, technologies, tools and software systems for efficient storage, processing, analysis, mining and visualization of Big mobility, transport and health-related data collected using mobile crowd sensing and Internet of Things paradigms in Smart Cities.

Researchers

Principle Investigator: Prof. Dragan Stojanovic || Contact Author: Prof. Dragan Stojanovic || University of Nils, Serbia

Neurocomputational Economic Forecasting in Turbulent Times

Abstract

Economic and business forecasting in turbulent times is hard, as data are nonstationary. Yet, the brain has mechanisms to deal with such challenges and neuroscience has uncovered some of them. Here we predict economic indicators by stochastically optimizing a neural circuit model for emotion dynamics, using only a handful of observations.

 

Researchers

Principle Investigator: Prof. Dr. George Mengov || Contact Author: Prof. Dr. George Mengov || Sofia University St. Kliment Ohridski, Bulgaria

Development of a probabilistic model for control the spread of infections on networks of contacts

Abstract

Contact tracing is a key element in countering an epidemic. We propose a probabilistic network model that by exploiting interactions between individuals infers their probability of being infected. This will enhance the information available and will enable the design of more focused restriction and control policies and evaluate their effects.

 

Researchers

Principle Investigator: Elisabetta Fersini || Contact Author: Elisabetta Fersini || University of Milan - Bicocca, Italy

FAST AND NON INVASIVE DIAGNOSIS OF SARS-COV-2 VIA RAMAN SPECTROSCOPY AND DEEP LEARNING

Abstract

Our preliminary work demonstrates the power of combining Raman spectroscopy and Deep Learning for a fast and non-invasive diagnosis of SARS-COV-2 infection from human salivary samples. In the next steps, we would like to investigate more complex DL architectures enabled by an increased data collection  to achieve proper clinical settings standards.

 

Researchers

Principle Investigator: Full Professor Vincenzina Messina || Contact Author: Dario Bertazioli || University of Milan - Bicocca, Italy

Development of deep-learning algorithms for reconstructing particle collisions at the LHCb detector

Abstract

The Deep-learning based Full Event Interpretation (DFEI) project is an ongoing initiative to increase the capabilities of the software trigger of the LHCb experiment at the LHC, by developing new deep-learning based algorithms. The huge complexity of the collision events requires the usage of HPC to allow the training of the algorithms.

 

Researchers

Principle Investigator: Prof. Marta Calvi || Contact Author: Ph.D. Julián García Pardiñas || University of Milano-Bicocca, Italy

WEEVIL- Eighth(WEEVIL8)

Abstract

The WEEVIL8 project is designed as the extension of the previous one (WEEVIL7), which was developed using the RX600S5-1 server from the HPI Future SOC Lab.  This project is a continuation of previous one, that is, this submission is a renewal for accessing HPI Future SOC Lab infrastructure.

 

Researchers

Principle Investigator: Prof. Dr. Carlos Juiz || Contact Author: Belen Bermejo || Universitat de les Illes Balears, Spain

Exploring Spiking Neural Networks for Real-Time Information Processing

Abstract

We want to examine a large artificial neural network in simulation of a system processing visual signals. Our first model was built of 4880 Hodgkin-Huxley neural cells. We now want to build a much larger 1 million-neuron system using a computational cluster and SAP HANA analytics platform for an in-memory processing of its readout signal.

 

Researchers

Principle Investigator: Dr. habil. (prof. PJAIT/UMCS) Grzegorz Marcin Wójcik || Contact Author: Karol Chlasta || Polish-Japanese Institute of Information Technology in Warsaw, Poland

Benchmarking Java on Ethernet Cluster

Abstract

The aim of this project is to check the scalability of parallel, network intensive microbenchmarks and application written in Java, using the APGAS library and the PCJ library, HPC Challenge 2014 award-winning Java library for high-performance parallel computing, on the 1000 Core Cluster - with high performance Ethernet interfaces.

Researchers

Principle Investigator: Dr. Marek Nowicki || Contact Author: Dr. Marek Nowicki || Nicolaus Copernicus University in Toruń, Poland

Worldwide

Adaptive learning with concept drift for intrusion detection

Abstract

The goal of this project is to develop a non-intrusive automatic data collection mechanism to collect images of coral reefs in the Vamizi Island, in the north of Mozambique. Object detection algorithms will be used in real-time to automatically photograph, detect and classify fish and other marine species that will pass by the cameras.

 

Researchers

Principle Investigator: Dr. Kuljit Kaur Chahal || Contact Author: Sugandh Seth || Guru Nanak Dev University, India

A world with X-road

Abstract

Our proposal consists of an X-Road instance functioning as the security layer in our network between our Machine Learning algorithm and the databases. We hope to achieve a way for government institutions to find areas of opportunity in the services they provide to the public and implement policies that will be more adequate to satisfy them.

 

Researchers

Principle Investigator: Rajesh Roshan Biswal Roshan Biswal || Contact Author: Rajesh Roshan Biswal Roshan Biswal || Instituto Tecnológico y de Estudios Superiores de Monterrey (ITESM), Mexico

Document Analysis for Cultural Heritage Data

Abstract

The Wildenstein Plattner Institute is undertaking a massive digitization project, with the goal to make millions of previously unpublished cultural heritage information available to the broader public. Because of the mass of information, it can only be processed using ML techniques like advanced OCR and image processing algorithms.

 

Researchers

Principle Investigator: Prof. Dr. Christoph Meinel || Contact Author: Hendrik Rätz || Wildenstein Plattner Institute Inc., United States

QSAR/QSPR model towards the prediction of novel fullerene derivatives as drug nanocarriers

Abstract

This project aims to formulate the Quantitative structure-property/activity relationship (QSPR/QSAR) model in terms of physicochemical properties, calculated by means of quantum mechanical methods, of C60 fullerene derivatives and doxorubicin. This drug has been widely used to treat breast cancer, isolated as well as carried with C60 fullerene.

 

Researchers

Principle Investigator: Dr. Alan Joel Miralrio Pineda || Contact Author: Dr. Alan Joel Miralrio Pineda || Instituto Tecnológico y de Estudios Superiores de Monterrey (ITESM), Mexico

Complementing and Substuting Classical Analysis with Deep Learning

Abstract

Almost any question that has been discussed in statistical science has its analogue in learning theory. Symbolic and numeric learning are currently dominated by deep neuronal networks. Now, a number of bioinformatics pipelines in clinical practice can be rethought, aiming at higher sensitivity, etc.

Researchers

Principle Investigator: Head of Platform, PhD Juan Jose Lozano || Contact Author: Dr. Julia Sidorova || Biomedical Research Networking Center in Hepatic and Digestive Diseases (CIBEREHD), Spain

OPTIMIZATION OF VIRTUAL MACHINE SCHEDULING IN DATA CENTERS

Abstract

Management of computers efficiently and to schedule the workloads is one of the visions in the Cloud Computing (CC) principles and paradigms. Having the hardware costs dropping, the individual computers, devices or in general nodes in a cloud environment are getting cheaper with lots of storage or computing available for in surplus. 

Researchers

Principle Investigator: Dr. Saleema J S || Contact Author: Mrs. Lakshmi Sankaran || Christ University, India

On Open Source Software metadata analysis for Classifying and Predicting Project Characteristics

Abstract

With the advent of machine and deep learning algorithms for classification and prediction, there is need of runtime infrastructure for building models from version data of Open Source Software (OSS) projects. Using FSOC infrastructure, we propose to build models for better understanding of the OSS process, project, and product characteristics.

Researchers

Principle Investigator: Dr. Kuljit Kaur Chahal || Contact Author: Dr. Kuljit Kaur Chahal || Guru Nanak Dev University, India