What IS IT ABOUT?
Big Data Frameworks
We are using SQL and No-SQL frameworks for Big Data Analytics, including stream- and batch-based networking data processing
Machine Learning
We are building and testing novel Machine-Learning based approaches for large-scale network measurements analytics
Benchmarking
We are conceiving new benchmarks for comparing the performance of big data stream analysis and big network traffic monitoring solutions
Network Monitoring
We are developing novel techniques for large-scale network traffic monitoring and analysis, including network security and anomaly detection
PROJECT PARTNERS
Big-DAMA Use cases
Cyber attackers are continuously looking for new ways to access valuable information and disrupt services. Big-DAMA builds and evaluates specific Big Data Analytics frameworks and techniques for the purpose of detection and characterization of network attacks, considering both on-line data analysis, as well as the
extraction of relevant patterns for data forensics analysis.
Despite the long literature and assorted list of proposed
systems for performing automatic detection and diagnosis
of anomalies in large-scale operational networks, Internet Service
Providers (ISPs) are still looking for a Holy Grail which might
effectively detect and diagnose the ever-growing number of
network traffic anomalies they face in their daily business. Big-DAMA is conceiving a novel framework for detection and diagnosis of network traffic anomalies.
-
Detection of 0-day attacks, based on unsupervised analysis techniques
-
On-line detection and classification of network and service anomalies
-
Internet performance evaluation through large-scale measurements analysis
-
Using machine learning techniques to monitor networks from a user-centric perspective
Big-DAMA Team
Pedro Casas
Team Leader