This page contains the datasets related to Peering testbed. This includes catchment distribution datasets in peering, attacks that we used to test the effectiveness of our playbook with Peering, and catchment stability using B-root catchments. The other datasets and tools are available in the following links:
This web page documents our datasets for measuring anycast agility against DDoS in Peering testbed. Our datasets are available upon request. We provide datasets about traffic estimation and distribution after BGP changes in testbeds, attack data from a DNS root server and from a national scrubbing service, and other data related to anycast catchment stability and load distribution.
We provide several datasets related to anycast catchment distribution after changes in the routing configurations. We announce different BGP configurations, run ping measurements to millions of responsive targets, and then capture the responses from these measurements. We provide the raw pcap files captured from these measurements, and parsed data files in human-readable format. We provide the following data files for different anycast and routing configurations. The README for these datasets is available here.
Path_Prepending_AMS,BOS,CNF-20200224: impact of path prepending on 2020-02-24 using AMS, BOS, CNF sites - contains datasets for both positive and negative prepending (upto 3 times)
Community_Strings_AMS,BOS,CNF-20200225: community strings (at AMS) on catchments for AMS, BOS, CNF on 2020-02-25 - contains several community strings configurations supported by the testbed
Path_Prepending_BOS,SLC,SEA2-20200228: impact of choosing a different set of sites (BOS, SEA and SLC) on 2020-02-28
Path_Prepending_AMS,BOS,CNF-20200224, Path_Prepending_AMS,BOS,CNF,SEA2,ATH-20200407, Path_Prepending_AMS,BOS,CNF,SEA2,ATH,ATL,SLC-20200408: impacts of changing the number of anycast sites from 2020-04-07 to 2020-04-10 - contains both positive and negative path prepending and community string configurations for 3 sites (AMS, BOS, CNF), 5 sites (AMS, BOS, CNF, SEA, ATH) and 7 sites (AMS, BOS, CNF, SEA, ATH, ATL, SLC)
Path_Poisoning_AMS,BOS,CNF-20210323: impact of path poisoning - poisoning tier-1 and other ASes both one step and multi-hops away
Path_Prepending_ATH,BOS,CNF-20200530: impact of path prepending in catchment distribution with a different combination of sites (ATH, BOS and CNF) on 2020-05-30
Path_Prepending_BOS,ATL,MSN-20200529: impact of path prepending in catchment distribution with BOS, ATL and MSN sites on 2020-05-29
Path_Prepending_AMS,BOS,CNF-20200224, Path_Prepending_AMS,BOS,CNF-20200409, No_Prepend_AMS,BOS,CNF-20200601 (considering the baseline cases): catchment distribution over time - this includes three snapshots from three different months (2020-02, 2020-04, and 2020-06)
To make the Verfploeter measurement, one needs to use a hitlist of target IPs. We utilize the following hitlist for our measurement.
ANT periodically updates the hitlist with new target IPs. To make a BGP playbook, we encourage the users to use the most recent hitlist.
We provide attack data for DDoS events collected from B-root and Dutch national scrubbing center. For B-root, we provide the raw pcap files along the message question format data. For Dutch national scrubbing center, we provide prefix-wise load at every 5 seconds. The READMEs for B-root attack events are available here. The README for other attacks is available here.
Dutch scrubbing center - 2021-08-22 - not provided due to privacy reason
Dutch scrubbing center - 2021-08-25 - not provided due to privacy reason
Dutch scrubbing center - 2021-08-27 - not provided due to privacy reason
An enterprise attack data - not provided due to privacy reason.
In this dataset, we provide B-root catchment distribution for a month. We provide a list of /24 prefixes, and their corresponding catchment in B-root in July 2021. The README for this dataset is available here. This dataset contains:
B-root catchment distribution for production prefix (ID 15001)
B-root catchment distribution for old prefix (ID 15000)
B-root catchment distribution for test prefix (ID 15002)
If you are using these datasets, please cite this work.