Monday 25.6

08:00 – 09:00Morning Coffee


09:00 – 09:15Welcome and Introduction


09:15 – 10:30Tutorial I (1) – Next-gen Network Telemetry is Within Your Packets: In-situ OAM

Frank Brockners, distinguished engineer at Cisco, with the support of Shwetha Bhandari, software engineer at Cisco and Leonardo Linguaglossa, postdoctoral fellow at Telecom ParisTech.

Abstract: In-situ Operations, Administration, and Maintenance (IOAM) is a paradigm shift in monitoring networks. IOAM provides information about *all* the user traffic on a per-packet level allowing network operators to mine entirely new insights. IOAM does so by recording operations and management related information as metadata within the user data packets while the packets traverse the network. IOAM adds a new category to the set of mechanisms for network monitoring: It complements network operations mechanisms which are polling based (ping, traceroute) or observation based (SNMP polling, streaming telemetry). IOAM information can either be accessed directly on the router or be available via Netflow/IPFIX.

The session will introduce IOAM as a technology and discuss a series of use-cases and deployment scenarios, ranging from proving in a cryptographically secure way that all packets traverse a specific path (“Proof of Transit”) to detecting and rapidly isolating faults, checking whether a SLA in an overlay network is really in place, to smart service selection for microservices. The technology discussion will be complemented by a set of demos (using the Vector-Packet-Processor VPP developed by the FD.io open source project, OpenDaylight Controller, etc.) which showcase this new technology at work. It will evolve into a hands-on lab session where we’ll get a live experience of IOAM at work.


10:30 – 11:00Coffee Break + Poster Session


11:00 – 12:30Tutorial I (2) – Next-gen Network Telemetry is Within Your Packets: In-situ OAM


12:30 – 14:00Lunch Break + Poster Session

14:00 – 15:30Tutorial II (1) – Welcome to BigData Zoo!: Open Source Big Data Tools Bestiary with a special visit to Spark

Joseph Allemandou, data engineer at Wikimedia Foundation.

Abstract: We’ll first take a guided tour of the main open source big data tools, explaining some of their differences and usages. We’ll then focus on Spark, state-of-the-art tool in the general-distributed-computing-engines space, giving an example of how it can be used to train machine learning models for a classical network security task.

 
 


15:30 – 16:00Coffee Break + Poster Session


16:00 – 17:30Tutorial II (2) – Welcome to BigData Zoo!


18:30 – 23:00Dinner in traditional beer garden Vienna-relaxed-style restaurant

 

Tuesday 26.6

08:00 – 09:00Morning Coffee


09:00 – 10:30Tutorial III (1) – From Packets to Knowledge: Applying Data Science Approaches to Large Scale Passive Measurements

Idilio Drago, assistant professor at Politecnico di Torino, with the support of Martino Trevisan, PhD student at Politecnico di Torino.

Abstract: The decentralized nature of the Internet makes it complex to be understood and managed. Network managers face challenging tasks when configuring and administrating the network, from the analysis of performance and availability issues, to the identification of malicious activity. All such tasks are performed relying on network measurements that inform managers about the state of the network.

In this lecture students will practice basic steps for the analysis of network measurements. Participants will engage on a mini machine-learning challenge: They will compete for the best performing algorithm for fingerprinting the usage of Internet services, evaluating the implications for users’ privacy. Students will profit from a state-of-the-art big data processing framework (Apache Spark) to process large-scale datasets of passive network measurements. By solving the challenge, students will travel over all stages for extracting knowledge from network measurements, face common challenges and pitfalls in traffic analysis, and get a better feeling on how big data approaches can help on the task.

slides

Update: Congratulations to Flavia Salutari, Andrian Putina and Matteo Romiti who won the challenge proposed during the tutorial!


10:30 – 11:00Coffee Break + Poster Session


11:00 – 12:30Tutorial III (2) – From Packets to Knowledge: Applying Data Science Approaches to Large Scale Passive Measurements


12:30 – 14:00Lunch Break

14:00 – 17:30TMA Experts Summit


18:00 – 22:00Conference Welcome Reception

The welcome reception takes place in the same building of the PhD school, at the Tech Gate Vienna tower terrace.