Banner Banner

Big Data Engineering


Prof. Dr. Matthias Böhm


Technische Universität Berlin
Ernst-Reuter-Platz 7, 10587 Berlin

Data Science Abstractions and Systems, Performance-Accuracy Tradeoffs in Data Science, Data Cleaning Pipelines and Optimization


The mission of the Big Data Engineering group, led by Prof. Dr. Matthias Böhm, is to simplify data science by providing high-level, data-science-centric abstractions and building systems and tools to execute these tasks in an efficient and scalable manner. The general research interests include the exploration of performance-accuracy tradeoffs, tooling (script generators, label generation, advisors, etc.), seamless data augmentation, cleaning, feature engineering, model debugging and deployment, cost-effective cloud deployments, advanced optimization techniques, adaptive data storage and indexing, and the exploitation of modern hardware. 

Current research focuses on:
•    Data Cleaning Pipelines: Automatic enumeration of data cleaning pipelines for target ML application, hyperparameter optimization of cleaning primitives.
•    Model Debugging: Finding the top-k data slices where a trained model underperforms, linear-algebra-based enumeration and pruning algorithms.
•    Fine-grained Lineage Tracing and Reuse: Fine-grained, multi-level lineage tracing for versioning and reuse, lineage deduplication, full and partial reuse of intermediates.
•    Federated Linear Algebra and Parameter Servers: ML model training on federated raw data without central data consolidation, plan generation under awareness of privacy constraints, federated linear algebra programs and parameter servers.
•    Workload-aware Data Reorganization: Compression under awareness of data and workload (linear algebra program) characteristics, asynchronous data reorganization in standing executors (e.g., at standing federated workers).
•    Code Generation for Heterogeneous HW: Extended operator fusion and code generation for GPUs and heterogeneous devices, including sparsity exploitation across operations.

June 13, 2024

BIFOLD Researchers receive three SIGMOD Awards

Each year SIGMOD conference awards are bestowed on researchers who have especially contributed to the field of data management. In 2024 BIFOLD researchers were honored to receive three awards.

June 09, 2024

BIFOLD at the 2024 ACM SIGMOD/PODS Conference

BIFOLD researchers presented four research papers, two demos, one workshop paper and were of a panel at the 2024 ACM SIGMOD/ PODS Conference in Santiago, Chile.

Orji Joseph / unsplash
April 22, 2024

“POLAR” lowers the adoption barrier for adaptive query processing in database systems

A preprint by BIFOLD researchers titled "POLAR: Adaptive and Non-invasive Join Order Selection via Plans of Least Resistance" is set to be presented at the VLDB conference in 2024. The database engineering paper introduces a technique for reordering joins that is adaptive, with a focus on non-invasive integration and low overhead. 

Prof. Dr. Matthias Böhm

Research Group Lead

Sebastian Baunsgaard Bifold researcher

Sebastian Baunsgaard

Doctoral Researcher

Dr. Patrick Damme Researcher BIFOLD

Dr. Patrick Damme

Postdoctoral Researcher

Sarah Hashmi

Secretary MLSEC & DAMS

David Justen

Doctoral Researcher

Arnab Phani BIFOLD researcher

Arnab Phani

Doctoral Researcher

Kindly take note that only researchers who have received funding from BIFOLD have their individual profiles displayed on