# OSPF Protocol Analysis: Historical Router CPU Performance and Data Center Scaling Claims

DevFeed: [OSPF Protocol Analysis: Historical Router CPU Performance and Data Center Scaling Claims](<https://devfeed.tech/articles/must-read-ospf-protocol-analysis-rfc-1245-11017.md>)

Original publisher: [Read original article](<https://blog.ipspace.net/2024/05/worth-reading-ospf-protocol-analysis/>)

Published: 2024-05-16T06:40:00Z

Content type: opinion

Language: en

Sources: [ipSpace.net blog](<https://devfeed.tech/sources/ipspace-net-blog.md>)

Topics: [networking](<https://devfeed.tech/topics/networking.md>), [Networks](<https://devfeed.tech/topics/networks.md>), [cpu](<https://devfeed.tech/topics/cpu.md>), [datacenter](<https://devfeed.tech/topics/datacenter.md>), [IS-IS](<https://devfeed.tech/topics/is-is.md>)

Tags: [architecture](<https://devfeed.tech/tags/architecture.md>), [cisco](<https://devfeed.tech/tags/cisco.md>), [cpu](<https://devfeed.tech/tags/cpu.md>), [data-center](<https://devfeed.tech/tags/data-center.md>), [intel](<https://devfeed.tech/tags/intel.md>), [is-is](<https://devfeed.tech/tags/is-is.md>), [mips](<https://devfeed.tech/tags/mips.md>), [model](<https://devfeed.tech/tags/model.md>), [ospf](<https://devfeed.tech/tags/ospf.md>), [protocol](<https://devfeed.tech/tags/protocol.md>), [router](<https://devfeed.tech/tags/router.md>), [time](<https://devfeed.tech/tags/time.md>)

## AI overview

This commentary reviews the Router CPU section of the 1991 OSPF Protocol Analysis (RFC 1245), including historical estimates for running Dijkstra's calculation and comparisons with modern CPU performance. It argues that CPU limitations alone do not justify replacing OSPF or IS-IS with EBGP for data center fabrics, while noting that implementation quality can matter.

## Source excerpt

Daniel Dib found the ancient OSPF Protocol Analysis (RFC 1245) that includes the Router CPU section. Please keep in mind the RFC was published in 1991 (35 years ago): Steve Deering presented results for the Dijkstra calculation in the "MOSPF meeting report" in [3]. Steve's calculation was done on a DEC 5000 (10 mips processor), using the Stanford internet as a model. His graphs are based on numbers of networks, not number of routers. However, if we extrapolate that the ratio of routers to networks remains the same, the time to run Dijkstra for 200 routers in Steve's implementation was around 15 milliseconds. Read more ...