# Stop the Network-Based Application Recognition Nonsense

DevFeed: [Stop the Network-Based Application Recognition Nonsense](<https://devfeed.tech/articles/stop-the-network-based-application-recognition-nonsense-10986.md>)

Original publisher: [Read original article](<https://blog.ipspace.net/2024/04/quic-nbar-ai-nonsense/>)

Published: 2024-04-24T06:35:00Z

Content type: opinion

Language: en

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

Topics: [Networks](<https://devfeed.tech/topics/networks.md>), [Cisco](<https://devfeed.tech/topics/cisco.md>), [Optimization](<https://devfeed.tech/topics/optimization.md>), [client](<https://devfeed.tech/topics/client.md>)

Tags: [cisco](<https://devfeed.tech/tags/cisco.md>), [network](<https://devfeed.tech/tags/network.md>), [opinion](<https://devfeed.tech/tags/opinion.md>), [optimization](<https://devfeed.tech/tags/optimization.md>), [qos](<https://devfeed.tech/tags/qos.md>), [traffic-engineering](<https://devfeed.tech/tags/traffic-engineering.md>)

## AI overview

The article argues that service providers should prioritize simple, fair congestion management based on individual-user link utilization instead of relying on application recognition and Cisco Ultra Traffic Optimization AI. It also places current concerns about QUIC in the context of earlier claims that UDP, BitTorrent, or video would overwhelm networks.

## Source excerpt

One of my readers sent me an interesting update on the post-QUIC round of NBAR whack-a-mole (TL&DR: everything is better with Bluetooth AI): Cloudflare (and the other hyperscalers) are full into QUIC, as it gives them lots of E2E control, taking a lot of choice away from the service providers on how they handle traffic and congestion. It is quite well outlined by Geoff Huston in an APNIC podcast. So far, so good. However, whenever there's a change, there's an opportunity for marketing FUD, coming from the usual direction. Read more ...