Berlin Institute for the Foundations of Learning and Data (BIFOLD)

© BIFOLD

The Berlin Institute for the Foundations of Learning and Data (BIFOLD), has evolved in 2019 from the merger of two national Artificial Intelligence Competence Centers: the Berlin Big Data Center (BBDC) and the Berlin Center for Machine Learning (BZML). Embedded in the vibrant Berlin metropolitan area, BIFOLD provides an outstanding scientific environment and numerous collaboration opportunities for national and international researchers. BIFOLD offers a broad range of research topics as well as a platform for interdisciplinary research and knowledge exchange with the sciences and humanities, industry, startups and society. more to: Berlin Institute for the Foundations of Learning and Data (BIFOLD)

Cheetah

In this project, we are implementing a set of operators for a prototype stream processing engine. To optimise performance, we work on heterogeneous hardware. To take advantage of the different hardware architectures, a universal software approach is developed for each operator and implemented across architectures. The implementation is followed by an analytical phase in which the operators are tested in different scenarios to find the parameterisation that maximises performance. By adapting the software approach to the hardware, we hope to achieve results that surpass the current state of the art. more to: Cheetah

DORIAN

As part of this project, we design the high-level abstraction for the declarative specification of the DS workflows. We implement a prototype of the management system that automatically extracts this declarative intermediate representation (IR) from a data science experiment and persists it in an experiment database for further reproducibility, search, comparison, and reuse. more to: DORIAN

ELEGANT

© ELEGANT project

ELEGANT aims to create a new software paradigm for Big Data and IoT by unifying their programming environments and enabling the automatic and easy deployment of existing code from Big Data platforms to IoT devices and backwards in a self-adaptable way, optimizing the PESRD (Performance, Energy Efficiency, Security, Reliability, Dependability) space. more to: ELEGANT

ExDRa

The Exploratory Data Science Over Raw Data (ExDRa) Project aims to conduct R&D in exploratory data science. The objective is to provide a support mechanism for exploratory data science and ease the analysis of distributed, heterogeneous raw data as well as to develop a research prototype suitable for real-world use cases. more to: ExDRa

FONDA: Debugging Distributed Data Analysis Workflows

The DFG Collaborative Research Center "FONDA - Foundations of Workflows for the Analysis of Big Data in the Natural Sciences" is dedicated to the optimization of data analysis workflows. The goal is to explore techniques, procedures and tools that enable an increase in the productivity of scientists in the creation and application of DAWs on large natural science datasets. more to: FONDA: Debugging Distributed Data Analysis Workflows

Hawk - A Hardware Adaptive Query Compiler

© CoGaDB Team

The performance of modern processors is primarily bound by a fixed energy budget. This power wall forces processor vendors to specialize their processors to certain applications to provide the speedups users expect. more to: Hawk - A Hardware Adaptive Query Compiler

moreEVS

The moreEVS Project is part of a bilateral initiative for joint Sino-German research projects. The number of electric vehicles (EVs) in urban areas, and therefore also the need for charging stations is expected to increase rapidly in the following years. In this research project, DIMA will address the challenge of conducting large-scale data analysis efficiently and pair renewable energy power sources with EVs. more to: moreEVS

SpaDa

In this project, we want to face these challenges to enable efficient processing of complex stream processing pipelines on modern hardware. To this end, we propose a novel adaptive query compiler for stream processing techniques to optimize code with regards to the hardware resources and changing data characteristics. Furthermore, we study possibilities to embed complex user-defined functions into compiled pipelines efficiently to support a wide range of advanced analytical data processing workloads. more to: SpaDa