Prepare List of Potential Project usecases based on Maggelan possiblities

[yadoshlivyy]

RIPE Atlas provides valuable data that can be used to research various network phenomena, including:

• Latency and packet loss: By using active measurements like ping and traceroute, researchers can monitor latency and packet loss between probes and other targets in the network, which helps in detecting network congestion and identifying routing issues.

• DNS performance and reachability: RIPE Atlas probes can perform DNS queries to monitor the performance and reachability of DNS resolvers and root servers. This helps researchers understand the global state of the Domain Name System and detect possible issues in name resolution.

• Routing stability: Researchers can use Border Gateway Protocol (BGP) data collected from RIPE Atlas anchors to monitor routing stability and identify BGP-related issues, such as route leaks, hijacks, and misconfigurations.

• IPv6 deployment and performance: RIPE Atlas enables researchers to study IPv6 adoption, connectivity, and performance by comparing measurements over IPv4 and IPv6.

• Network topology: By collecting and analyzing traceroute data, researchers can understand the underlying structure of the Internet, identify points of centralization, and detect potential bottlenecks.

• Content Delivery Networks (CDNs) and peering: Researchers can study the performance of CDNs and the effects of peering agreements by monitoring the paths taken by traffic between RIPE Atlas probes and popular Internet services.

• Network resilience and security: RIPE Atlas can be used to monitor the impact of Distributed Denial of Service (DDoS) attacks, routing incidents, or other network disruptions, and to assess the resilience of the Internet infrastructure.

• Internet Exchange Points (IXPs): Researchers can use RIPE Atlas data to study the role of IXPs in the global Internet infrastructure and to understand their impact on latency, routing efficiency, and overall network performance.

• Time synchronization and network time: By monitoring the performance of Network Time Protocol (NTP) servers, researchers can study time synchronization issues in the Internet and identify potential problems.

These are just a few examples of the many network phenomena that can be researched using RIPE Atlas data. The versatility of the platform makes it a valuable resource for network operators, researchers, and policymakers interested in understanding and improving the global Internet infrastructure.

[levitanda]

How do sanctions affect the Internet?

• Computers on the Internet address each other through long strings of numbers that uniquely identify points from which packets can be sent and received. These numbers - Internet Protocol (IP) addresses - are distributed by the Regional Internet Registries (RIRs), usually to Internet Service Providers who then allocate them, in turn, to their customers. The RIRs also distribute the Autonomous System numbers (ASNs) that uniquely identify networks on the Internet.

Policies, regulations, and technical problems that curtail the distribution of IP addresses can thus have a significant impact on user access. Rather than merely depriving users of access to specific online services, lack of access to IP addresses effectively prevents people from going online at all.

• Inequitable access to number resources There may be inequitable access to number resources - the IP addresses and ASNs mentioned above - because of the indirect consequences of sanctions. Using compliance processes, RIRs can do the minimum to stay in compliance with sanction regimes and provide their services legally to the sanctioned countries. However, there is only so much they can do compliance-wise. When the impact of sanctions is indirect and entangled with other industries, then inequitable access to number resources might emerge.

The inequitable access can happen because nationals of sanctioned countries that are not themselves the target of sanction regimes might not be able to register new number resources. After all, banks are not willing to facilitate their transactions. The issue goes beyond that; sometimes, countries that are not sanctioned but are transacting and sharing IP blocks with sanctioned countries might be affected as well.

• When specific sanctions apply to individuals with formal roles in telecommunication services (for example, the CEO of a telecom operator), the Internet Exchange Point that is subject to the sanctions regime in question will have to terminate the network operator’s membership. This‌ can have the following consequences:

De-peering has consistently been recognised as an extreme step, as it means customers might not reach specific sites on the Internet. (See this paper for more information.) If the network operator is large and serves smaller network operators, those network operators are also affected. This will affect the quality of access and create latency. Some argue (as reported in Russian state-owned media) that it does not impact their services. Such network operators claim they can have access to global traffic through Asia. But there are restrictions. For example, it is difficult to peer with Chinese operators due to their domestic restrictions on Internet traffic. Network operators that are sanctioned might carry Internet traffic of other non-sanctioned countries. In such a case, the sanctions (and revocation of membership from IXPs) can affect other network operators based in other countries. When revocation of membership from a well-established Internet Exchange Point happens, the individual members of that exchange point will likely stop peering with the sanctioned network bilaterally.

• Cache servers are a means of making sure that much of the most popular content available on the Internet is always “close”, in a network sense, to the end user. These services enable the web, in particular, to satisfy enormous demands. Cache servers do not necessarily serve a peering function, but they are essential for cloud providers and peering locations, as well as for the quality of access to the Internet. They are even sometimes critical for having meaningful access to the Internet. A cache server temporarily stores information on a local network, making browsing faster. Cache servers are usually installed in data centres, ISPs, and peering locations. Trade restrictions, export, and import controls, and sanctions could impact the availability of these servers. There were two reported cases of Google shutting down its caching servers in two Russian ISPs. Google (reportedly) stated that the reason was a change in legal practices and compliance with sanctions. There are reports about Cache servers being unavailable in Afghanistan as well.

https://labs.ripe.net/author/farzaneh-badiei/how-do-sanctions-impact-the-internet/

[levitanda]

So what we can observe in the project??

1. Using BGP data from RIS, we look at how networks in any country depend on each other and on foreign networks. We take all ASNs for the given country, and use RIS to find all the direct links these ASNs have - either between two ASNs for the given country, or where one of the ASNs depends on a foreign ASN. Of these links, we only look at those where there is a strong dependency of one network upon the other, as determined by AS-Hegemony. GOOD RESOURSE

2. Minimum RTT - We can extract the minimum round trip time (RTT) that we see in traceroutes from each RIPE Atlas probe into each network that we have collected traceroute data for on a given day. The network data is aggregated by their well-known identifier, the Autonomous System Number (ASN), but we also aggregate data for IXP peering LANs in as far as they are recorded in PeeringDB. So for a network or IXP, this dataset will contain a single minimum RTT value for all RIPE Atlas probes that see IP addresses for that particular network or IXP for a given day. And this requires no extra measurement as it builds on top of measurements that already exist in RIPE Atlas. The resulting dataset is an interesting aggregate. Not only is it small (500MB uncompressed per day), but as we'll show below, it can provide insights for the following use-cases:\ What is the latency into a network as seen from over 10,000 vantage points all over the world? Which networks are close/far from a particular vantage point? There is created prototype visualisations, using the ObservableHQ platform. This allows for rapid and agile prototyping of data ideas. If you have programming and/or visualisation skills, you can easily fork this code and create your own version of it and, if you want, contribute back your improvements.

3. Latency map, shows the lowest latency into particular networks or IXPs as seen from RIPE Atlas, so not only for a particular IP end point. https://labs.ripe.net/author/emileaben/latency-into-your-network-as-seen-from-ripe-atlas/ - Here you can find useful examples https://observablehq.com/@ripencc/atlas-latency-worldmap

4. Zombie BGP route - A BGP zombie refers to an active routing table entry for a prefix that has been withdrawn by its origin network, and is hence not reachable anymore. This issue can affect the reachability of the internet. There is an hypotise: "zombies are mainly the results of software bugs in routers, BGP optimisers, and route reflectors", and we can check it. More about this case - https://labs.ripe.net/author/romain_fontugne/bgp-zombies/

*Great resourse for graphs drawing https://gephi.org , example for using - https://labs.ripe.net/author/emileaben/the-resilience-of-the-internet-in-ukraine-one-year-on/

*A lot of sources use AS, IXP and country name filters. Maybe we should too.

[yadoshlivyy]

1) HIgher DNS RTT -> the more congested network. 2) Bigger amount of hops in trace route -> the more congested network

Detecting high network congestion using RIPE Atlas data involves analyzing key metrics from different types of measurements, such as ping, traceroute, and DNS. These measurements can help you identify latency, packet loss, and other indicators of network congestion. Here's how to use RIPE Atlas data to detect network congestion:

1. Latency (RTT) analysis: High latency or significant fluctuations in latency can indicate network congestion. Analyze ping measurement data to calculate the average Round Trip Time (RTT) for packets between RIPE Atlas probes and specific targets. Observe latency trends over time to identify periods with high or varying RTT values, which might suggest congestion.

2. Packet loss analysis: Packet loss is another critical indicator of network congestion. Using ping measurement data, calculate the percentage of lost packets between RIPE Atlas probes and targets. High packet loss rates can point to network congestion or other connectivity issues.

3. Traceroute analysis: Analyze traceroute measurements to identify specific network segments or routers experiencing high latency or packet loss. This can help pinpoint the locations of network congestion. Additionally, look for routing issues, such as routing loops or suboptimal routing, which can contribute to congestion.

4. DNS performance: Analyze DNS measurement data to assess the performance of DNS resolvers and root servers. High DNS response times or low success rates can indicate congestion or other network issues affecting the DNS infrastructure.

5. Regional analysis: Aggregate the measurement data by geographical regions or Autonomous Systems (ASes) to identify areas experiencing higher latency, packet loss, or other signs of network congestion.

To detect high network congestion using RIPE Atlas data, follow these steps:

1. Obtain the relevant RIPE Atlas measurement data using the API or data archives for the time period and targets of interest.
2. Preprocess and analyze the data to calculate key metrics, such as average RTT, packet loss rates, and DNS performance.
3. Visualize the metrics over time to identify trends and anomalies that might indicate network congestion.
4. Examine traceroute data to identify specific network segments or routers experiencing congestion or other issues.
5. Aggregate the data by regions or ASes to pinpoint areas most affected by network congestion.

By using RIPE Atlas data to analyze and monitor key network performance metrics, you can effectively detect high network congestion and take appropriate actions to mitigate its impact on your network or services.