Why I Think Time Sensitive Networking (TSN) Is Silly and Dangerous

แชร์
ฝัง
  • เผยแพร่เมื่อ 20 ก.ค. 2024
  • I may not be around much longer after posting this personal opinion on Time Sensitive Networking (TSN). There is a whole contingent of folks around the world that see TSN as one of the greatest inventions ever.
    I DON’T AGREE. I see it as silly and dangerous!
    Watch the video. I’ll be careful about starting my car - you know what I mean if you watched the Godfather.
    As always you can contact me here: www.rtautomation.com/contact/ or on LinkedIn: linkedin.com/in/johnsrinaldi
  • วิทยาศาสตร์และเทคโนโลยี

ความคิดเห็น • 11

  • @vonnikon
    @vonnikon ปีที่แล้ว +14

    802.1Q is not TSN.
    And TSN is not just 802.1Qbv.
    And TSN 802.1Qbu is a nice backwards compatible QoS mechanism without any disadvantages.
    And mixing IT traffic with OT in the same network is not a TSN issue: You can mix IT with OT without TSN, and you can use TSN without mixing IT with OT.
    Video is getting a lot of things fundamentally wrong.

    • @peqoudaut
      @peqoudaut ปีที่แล้ว +1

      This answer should be voted up. I could not see any point in his list that shows that this is "dangerous".
      I would agree that TSN might not make sense in a lot of use-cases, but at least in vehicles it really helps!

    • @Keith-ej1sx
      @Keith-ej1sx ปีที่แล้ว

      Agreed!

  • @magoostus
    @magoostus 5 หลายเดือนก่อน

    So I come from the audio industry, and I'm happy to see what other uses TSN has besides audio. I agree with some of your points, that in industrial automation, there might be cases that will not have a benefit from the added complexity that TSN brings, but also TSN has so *many* uses (eg realtime audio and video) that it won't be going away, it's going to be a fact of life. also IEEE802.1CB specifies redundancy, ring networks are supported, multiple paths are supported

  • @wildLeaner
    @wildLeaner 2 ปีที่แล้ว +2

    I like the way you critic, thank you, you gave much information!

  • @leesmithwigan1985
    @leesmithwigan1985 2 ปีที่แล้ว +1

    Sitting on the fence again I see John! 🤣

  • @viennabrunn
    @viennabrunn ปีที่แล้ว

    The argumentation is different if someone wants to have a virtualized embedded cloud hosting many hard RT, real-time and soft-time functions on few computers. Yes, the methodology, design tools and SW/HW components are important for safety-/mission-/time-critical functions.
    TSN standard makes more sense in cars (Nx100M of switches, >1Billions endpoints), eventually in the edge computing/factory automation. In the fieldbus domain it can add value to some of the existing industrial Ethernet standards by providing real-time capabilities.

  • @1durch0
    @1durch0 2 ปีที่แล้ว +1

    Good points

  • @nataschabaunse9151
    @nataschabaunse9151 10 หลายเดือนก่อน

    The scheduler will be some AI like Bard in cooperation with quamtum computing; it´s beeing build, it just takes time! But yes, I agree in one point with you - linking IT to OT - I don´t see the reason why either to risk OT to hacker!

  • @weitsai7387
    @weitsai7387 ปีที่แล้ว

    Great! Totally agreed!

  • @bobtraskpe
    @bobtraskpe ปีที่แล้ว +1

    Listen to this man.