Id |
Project |
Description |
Link |
Students |
1 |
P4 load-balancing |
Demonstrate the implementation of
P4 load-balancing in a configuration with 5 nodes and 3 links per node |
https://github.com/p4lang/tutorials/tree/master/exercises/load_balance |
Baptiste Thomas. |
2 |
P4 in-network computing |
Implement an in-network computing
system based on the BMv2 software switch. Demonstrate the execution of at
least 2 instructions. |
Paper: Ganesh C. Sankaran, Krishna M. Sivalingam,
Harsh Gondaliya, "P4 and NetFPGA
based secure in-network computing architecture for AI-enabled Industrial
Internet of Things" |
|
3 |
P4 heavy-hitter detection |
Demonstrate the implementation of
a heavy hitter detection using P4, showing how the detection performance
decreases as the traffic increases |
https://github.com/nsg-ethz/p4-learning/tree/master/exercises/06-Heavy_Hitter_Detector |
|
4 |
P4 fast-reroute |
Demonstrate the implementation of
fast rerouting with loop free alternative with P4, in a configuration with 6
nodes and 4 links per node. |
https://github.com/nsg-ethz/p4-learning/tree/master/exercises/12-Fast-Reroute |
BERGER Charles BOIVIN Benoit |
5 |
Container auto scaling-1 |
Implementation of vertical autoscaling of docker
containers/pods (prepare a lab excercise) |
https://www.giantswarm.io/blog/vertical-autoscaling-in-kubernetes |
|
6 |
Container auto scaling-2 |
Implementation of horizontal autoscaling of
docker containers/pods (prepare a lab excercise) |
https://kubernetes.io/docs/tasks/run-application/horizontal-pod-autoscale/ |
Vincent Thibaut |
7 |
Real-world Dataset
de-anonymization |
Exploit the analysis of crafted
malicious packets in a real-world dataset (2022) to de-anonymize it. Apply it
to at least one trace per week for the year 2022 (e.g., every Monday) |
Paper: will be provided by email
(in French)
|
Gauvain CRAHE |
8 |
Anomaly detection with ISF algorithm |
Determine the performance of Isolation forest to detect anomalie/outliers
in time-series data |
|
|
9 |
Anomaly detection with
transformers |
Determine the performance of
transformers to detect anomalie/outliers in
time-series data |
|
|
10 |
Anomaly detection with GRU |
Determine the performance of Gated
recurrent unit (GRU) to detect anomalie/outliers in
time-series data |
Dufort Ornecipe |
|
11 |
P4-Runtime based controller for
fast failure recovery |
Demonstrate the application of a
novel control plane from recent scientific literature to achieve network
failure detection and recovery |
Paper: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10036033 |
|
12 |
OVS vs bmv2 software switches |
Compare the performance of the two
software switches in terms of switching latency and packets per second with
different bitrates |
OVS: https://docs.openvswitch.org/en/latest/ |
Oumar Ndongo |
13 |
Container scaling for IoT video
traffic |
Assess the accuracy and the
throughput of an AI model and the network utilization under varying arrival
rate of networked IoT video streams. Consider for a single AI model (yolov8)
3 cases: (1) AI model deployed in a single container of medium size; (2) AI
model replicated in 5 containers of small size; (3) AI model deployed in a
single container of size equals 5 times the small size. |
https://docs.ultralytics.com/modes/ |
VISVAL
Thierry |
14 |
Container Migration |
Implement and demonstrate
migration of a container between 2 servers |
Bououf Faiçal Lavaury Yannick |
|
15 |
Universal Packet Scheduling |
Demonstrate the capability of
Least Slack Time First (LSTF) scheduling algorithm to minimize Mean Flow
Completion Time and Tail Packet Delay, as well as assuring fairness. |
Paper: github: |
Gérard Delorme |
16 |
Mirai Botnet DDoS Attack |
Deploy a Mirai Botnet testbed using
containers, consisting of one CnC (Command and
Control), one receiver, two bots, and one target host. The objective is to
showcase at least one of Mirai's fundamental
functions, including scanning, infection, and DDoS attacks. |
|
|
17 |
Service Function Chain (SFC)
traffic scheduling simulator |
Demonstrate the use of the SFC TSS
discrete event simulator |
|
|
18 |
eBPF |
Demonstrate at least three eBPF applications |
|
Mame DIAGNE |
19 |
Deep Learning for LDPC decoding in
wireless connections |
Compare a Deep Learning based
solution for Low-Density Parity-Check decoding with the conventional Min-Sum
decoder in terms of Bit Error Rate under different Signal-to-Noise ratio |
Paper: Github: |
Dufort ORNECIPE |
20 |
SniffnDetect |
Demonstrate the use of the Scapy-based DDoS detector SniffnDetect
2.0 to detect at least 3 kinds of attacks and evaluate its resource usage in
terms of RAM and CPU under different conditions (idle, during attacks, during
attacker identification) |
Sadki Abderrahil |
|
21 |
Network attacks with Scapy |
Demonstrate the use of Scapy to perform Wifi deauthentication, ARP cache poisoning and DHCP Starvation
attacks. Use the code from the suggested github
repos (or more) to create a single platform to launch several attacks. |
https://github.com/veerendra2/wifi-deauth-attack |
REIST Clément |
22 |
Snort Intrusion Detection |
Demonstrate the use of Snort3 as
Intrusion Detection System analysing PCAP files including TCP SYS scans and
DoS attacks |
https://github.com/snort3/snort3_demo/tree/master |
JOSEPH-MONDESIR Gérald |
23 |
Labelled data vs unlabelled data |
Use ISF ML algorithm and compare
the ML performance metrics |
https://github.com/mhdadizadeh/CICDDoS2019/blob/master/CICDDoS2019.ipynb |
|
24 |
Binary neural networks |
Compare the performance of BNNs to
other non BNN models (choose another DL model) |
https://github.com/mhdadizadeh/CICDDoS2019/blob/master/CICDDoS2019.ipynb https://arxiv.org/abs/2110.06804 |
|
25 |
Rancher demo |
Demo Rancher by deploying and managing
a microservices-based application. The project will demonstrate how Rancher
simplifies the deployment and orchestration of containerized applications. |
https://ranchermanager.docs.rancher.com/pages-for-subheaders/quick-start-guides |
Julien Dillenseger |
26 |
KubeVirt |
Illustrate the capacity of KubeVirt by demonstrating how it can create and manage virtual
machines (VMs) within a Kubernetes environment. |
Freddy PEZERON |
|
27 |
Firewall BPF/XDP |
Simulate a denial
of service attack and demonstrate the performance of BPF/XDP in
handling and mitigating the attack by efficiently filtering and processing
network traffic. |
http://arthurchiao.art/blog/firewalling-with-bpf-xdp/ https://dev.to/xenbytes/simple-xdp-firewall-with-golang-1da3 |
Othmane Essabil |
28 |
eBPF exploitation |
Demonstrate some security
vulnerabilities in eBPF and effective mitigations
to enhance the security of eBPF-enabled
environments. |
KEFKEF Ahmed YILDIRIM Umit |