Freigeben über


Quality of Experience (QoE) paper published

Tom has posted about the QoE paper that has been published. Please read it at: https://blogs.technet.com/toml/archive/2007/07/27/quality-of-experience-qoe-paper-published.aspx?CommentPosted=true#commentmessage

Comments

  • Anonymous
    January 01, 2003
    Hi Chris, I understand your comments about QoS v. QoE. There is a difference between the two. Simply put, QoS only focuses on the network and its relative health. Cisco and the other network infrastructure vendors have done a very good job at promoting QoS as a means to deliver "valuable" packets to their destination. However, QoS does not address the problem of what happens if the destination isn't ready for the packet. I wouldn't expect these vendors to address this either. So, this isn't a dig at them. They control the network and they are doing everything within their power to make the network the best it can be. QoE on the other hand takes into account not only the network, but processor consumption, memory usage, disk utilization. In other words, QoE is an attempt to look at any bottleneck within the infrastructure that may affect the quality of hte voice taffic.

  • Anonymous
    September 05, 2007
    Hi Chad Beleive it or not, I've been waiting for this to come out.  I only wish one less acronym had been introduced and a consistency with QoS had been established.   I'm still no wiser whether QoE is a lower or higher benchmark, or a completely different bench. Hope I'm not killing the messenger here, but I think QoE is one step too far.

  • Anonymous
    September 12, 2007
    Thanks for the explanation for that Chad.  I see what you are saying.  My feeling at the moment is that router and firewall vendors in general publish quite confusing documentation in their manuals regarding QoS.  This was a big headache for me.  Getting clear and simple QoE info in vendor manuals would be a major plus.

  • Anonymous
    November 27, 2007
    My client is looking delploy this in their enterprise, but are concerned about the overhead that the QMS may introduce into the network.  Are there any stats on how much load, bandwidth, etc. will be placed on the network as it monitors QoE?  We certainly don't want to degrade the QoE by measuring it!