ESCALATE: Efficient and Scalable Algorithms for Large Flow Detection

In the ESCALATE project, new algorithms will be developed and implemented for the detection of large data flows. Existing algorithms are effective and efficient for the detection of very large data flows (with a bandwidth that is, e.g., 100x larger than the allowed bandwidth). When the attack is mounted by several computers at the same time, it is, however, important to detect and block data flows that only exceed the allowed bandwidth to a limited extent. In ESCALATE, researchers of ETH Zürich will concentrate on algorithms that are effective and efficient for very large data flows as well as data flows that only exceed the allowed bandwidth by a factor 1.5. Because the speed of network devices is of ultimate importance, researchers of KU Leuven will design and implement new architectures for configurable hardware with the goal of integrating the configurable hardware in network devices. This way, network devices can be equipped with efficient and effective protection against large flow network attacks