Insights from our top experts about Technology

Using TagSniff for online and post-hoc data debugging

TagSniff is a data flow instrumentation approach that provides primitives for online and post-hoc debugging tasks, such as locating and fixing errors in computer programs and identifying tuples that need further analysis based on their metadata.

Author:

Dr. Jorge Quiané

The "TagSniff" Debugging Model

Application debugging is an essential task in big data engineering. We introduce the "TagSniff" model, a debugging abstraction that provides a streamlined set of instrumentations for easy and effective online and post-hoc debugging tasks.

Author:

Dr. Jorge Quiané

The need for Big Data Debugging

Big data processing is a growing field that requires new approaches to debugging complex, distributed data processing systems. Traditional debugging approaches are inadequate for the scale and complexity of big data systems, making them even more complex.

Author:

Dr. Jorge Quiané

Poisoning attacks in Federated Learning

Federated learning is a double-edged sword. It's designed to ensure data privacy, but unfortunately, it opens the door for adversaries to exploit the system by using poisoned data.

Author:

Dr. Euijin Alley Choo

The Missing Piece in Learning-based Query Optimization

In this blog post, we focus on using an ML in a place of a cost model and go from cost-based query optimization to the newly proposed ML-based query optimization.

Author:

Dr. Zoi Kaoudi

Towards a Learning-based Query Optimizer

Query optimization is at the core of any data management/analytics system. New research has used machine learning (ML) to solve these tasks more effectively and efficiently.

Author:

Dr. Zoi Kaoudi

Machine Learning for X and X for Machine Learning

In this blog post, we focus on big data management. We have already observed ML for data management as well as data management for ML over the last several years.

Author:

Dr. Zoi Kaoudi

Subscribe to our newsletter