We're now going to talk about efforts to improve aggregation using the CIDR report. The internet routing table as of August 2017 is shown on the screen. As you can see, there are 657,966 BGP routing table entries. Whereas the number of prefixes that would have been seen, had maximum aggregation been used, is 256,321. Bear in mind that maximum aggregation is calculated by Origin AS. This means that more than half of the prefixes in the routing table might be there unnecessarily. If you look at the unique prefixes we have 318,644 entries, this means operators are announcing prefixes from their blocks without a covering aggregate. Efforts to improve aggregation have been made possible by the creation of the CIDR report. This was initiated and operated for many years by Tony Bates in 1994. This has now been combined with Geoff Huston's routing analysis in the early 2000s, which can be found on www.cidr-report.org. This covers both IPv4 and IPv6 BGP tables. Results are emailed on a weekly basis to most operation lists around the world. The interesting bit is, the top 30 service providers who could do better at aggregation are listed and this could be embarrassing. RIPE Routing Working Group aggregation recommendations can be found on the URLs shown on the screen. The CIDR report also computes the size of the routing table assuming ISPs perform optimal aggregation. The website also allows searches and computations of aggregations to be made as per AS basis. It is a flexible and a powerful tool to aid ISPs. It is also intended to show greater efficiency in terms of BGP table size that can be obtained without loss of routing and policy information. It shows what forms of Origin AS aggregation could be performed and the potential benefit of such actions to the total table size. It very effectively challenges the traffic engineering excuse. These following screenshots show you what the CIDR report website looks like. As you can see, it shows you a status summary providing a table history of prefixes, aggregated prefixes, as well as summaries based on dates and corresponding graphs. It shows you the announced prefixes by AS, AS path, as well as the ranking. As you can see it will indicate the number of aggregates available in the AS, as well as suggest a number of prefix announcements that can be reduced. The non-aggregated prefixes are displayed in red, also indicating matching aggregate and a duplicate AS path. Properly generated sub prefixes are displayed in green. This detailed analysis helps operators in figuring out where to aggregate on their networks. We're now going to talk about the importance of aggregation. Aggregation helps in reducing size of the routing table. Router memory is not so much of a problem as it was in the 90s, as routers routinely carry over a million prefixes. However, when it comes to convergence of the routing system this is a big problem. Bigger table takes longer for CPUs to process. BGP updates also take longer to deal with. We now have a BGP instability report that tracks routing system update activity, this can be found on the URL shown on your screen. We're now going to talk about the convergence of the routing system and route instability. As you can see from the screenshot of the BGP instability report, more than 30 percent of BGP updates are generated by only 10 ASNs. The top ASN on the report sent more than 400,000 BGP updates in a week, which translates to more than 2,000 BGP updates in an hour and 40 BGP updates a minute. This is pretty excessive and it seems like more needs to be done to get worse culprits from the list, by encouraging them to get more stable networks as well as more aggregation. Apart from the top 10 ASNs in the report, the rest of them are sending no more than one BGP update a minute. If you look on the same report based on prefixes, you would expect to find /24s which is typically of customer networks flapping as demonstrated in the bad example previously. However, you see that there are a couple of /19s, /21s, and networks that are greater than /23 on the list. It is obvious from the list that aggregation is not being done by their service providers as several of them belong to the same ASN. As discussed previously, route damping might be done by other ISPs for these prefixes and QoS on these networks will definitely be bad. In conclusion, aggregation is the key to greater internet stability.
© Produced by Philip Smith and the Network Startup Resource Center, through the University of Oregon.
Attribution-NonCommercial 4.0 International (CC BY-NC 4.0)
This is a human-readable summary of (and not a substitute for) the license. Disclaimer. You are free to: Share — copy and redistribute the material in any medium or format Adapt — remix, transform, and build upon the material The licensor cannot revoke these freedoms as long as you follow the license terms. Under the following terms: Attribution — You must give appropriate credit, provide a link to the license, and indicate if changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use. NonCommercial — You may not use the material for commercial purposes. No additional restrictions — You may not apply legal terms or technological measures that legally restrict others from doing anything the license permits.