Interesting People mailing list archives
Re: worth reading -- QOS Author is Motorola, Chief Software Architect
From: David Farber <dave () farber net>
Date: Tue, 24 Jun 2008 15:52:51 -0700
________________________________________ From: Waclawsky John-A52165 [jgw () motorola com] Sent: Tuesday, June 24, 2008 6:39 PM To: David Farber; ip Subject: RE: [IP] worth reading -- QOS Author is Motorola, Chief Software Architect My intent with the "pipes comment" was not to deflect from the major points outlined further in my QoS observations note, that QoS is really very hard. I don't see it as an adequate substitute for capacity (my experiences were outlined). My intent was to use the "pipes" as way to express the fact that core network capacity is increasing and there is a good deal of capacity for various reasons related to design, reliability etc. because of not only pipe size but other factors (from Moore's law) such a node processing speed and memory capability, etc. Well run networks seem to be typically driven by network design objectives, good capacity planning, ability to support peak loads, etc. for connectivity, back-up and recovery and reliability etc.. I have found network have 4 kinds of delays: transmission or encoding delay (measured link by link, which is about putting packets of a certain size on a link of a certain speed), processing delay (time where packets wait at a node while it figures out what to do with or where to switch each packet), propagation delay (speed of light limitation over distance) and queuing delay (unproductive delay waiting somewhere in the network for a resource). I have found that network capacity planning/optimization is a never ending bottleneck resolution process. Remove one bottleneck and you find the next one. Certainly fiber offers enormous capacity and drives down one type of delay (such as encoding delay) causing another to potentially become a bottleneck focus (processing delay). Capacity from both pipes and nodes needs to be considered as part of good capacity planning practices and network design goals to keep traffic out of unproductive queues, which if packets wait too long will disturb some applications. But, regardless of where the bottlenecks are, getting all these queues and resources to operate under some end-to-end QoS scheme seems more than a little unlikely especially since QoS has been discussed since the dawn of networking ...and now adding node queues to the mix of things that need to be QoS managed doesn't appear to simplify the QoS challenges. The QoS ROI seems highly questionable and QoS brings in other issues, for example it is another thing to manage or configure incorrectly etc. (more in my original note) and QoS is NOT an adequate substitute for capacity in my experience. My 2 cents. -----Original Message----- From: David Farber [mailto:dave () farber net] Sent: Tuesday, June 24, 2008 10:15 AM To: ip Subject: [IP] worth reading -- QOS Author is Motorola, Chief Software Architect ________________________________________ From: Dave Crocker [dhc2 () dcrocker net] Sent: Tuesday, June 24, 2008 11:07 AM To: David Farber Cc: ip; Waclawsky John-A52165 Subject: Re: [IP] QOS Author is Motorola, Chief Software Architect David Farber wrote:
From: Waclawsky John-A52165 [jgw () motorola com] Sent: Monday, June 23,
2008
1:08 AM To: David Farber Subject: RE: [IP] Re: Net Neutrality: A
Radical
Form of Non-Discrimination by Hal Singer Hi Dave, Some QoS perspectives that I have learned: First, the main
problem.
QoS really isn't needed when you have big pipes.
This view has gained popularity in recent years and it seems to be based on two misunderstandings. The first is that end-to-end performance is dictated by the size of pipes and the second is that pipes are always large or that we can guarantee that eventually they all will be large. Packet switching is more about the switching than the pipes. The path from a one random end-system to another has quite a few switching points. This thing called queuing comes into play when there is transient contention for resources. This includes contention for use of each pipe along the way, but also contention in switches and, by the way, contention in either of the end-systems. (I'm qualifying with "transient" because sustained contention means that the system is fundamentally overloaded; queuing can't help there.) The premise behind the "big pipes" view is that we don't have transient contention. It's simply not true. What is true is that there are common scenarios where transient contention is almost never a problem. But the difference between "almost never" and "never" counts for everything in a world seeking reliability. Especially if you want to cover a full range of scenarios. One set of scenarios left out by "big pipe" devotees is a vast portion of the world with limited resources. While this obviously includes many remote or developing environments, it also includes less-capable channels such as mobile devices. It should also be noted that there is a tendency for the core of the Internet to have less contention than access networks at the edge. We can wave our hands and say that the edges will eventually catch up, but history suggests otherwise. A persistent lesson over the history of packet switching is that there is a wide range of resource capabilities and anything designed to rely on high-end capabilities disenfranchises participants and systems that are not so privileged. "QOS" has indeed had a problematic history over the life of packet-switching, but this seems to be because it is difficult to design in a way that is useful -- and then deploy it throughout the infrastructure -- rather than because it isn't needed. Basic Internet capabilities were designed to maximize use of the channels, but at the cost of inter-packet arrival variance. Any application needing to sustain a specific transmission rate with specific (and low) variance is at risk, without some underlying design to ensure the necessary performance. Anyone with experience to the contrary might want to review their sampling methodology against the full and realistic set of Internet scenarios. d/ -- Dave Crocker Brandenburg InternetWorking bbiw.net ------------------------------------------- Archives: http://www.listbox.com/member/archive/247/=now RSS Feed: http://www.listbox.com/member/archive/rss/247/ Powered by Listbox: http://www.listbox.com ------------------------------------------- Archives: http://www.listbox.com/member/archive/247/=now RSS Feed: http://www.listbox.com/member/archive/rss/247/ Powered by Listbox: http://www.listbox.com
Current thread:
- worth reading -- QOS Author is Motorola, Chief Software Architect David Farber (Jun 24)
- <Possible follow-ups>
- Re: worth reading -- QOS Author is Motorola, Chief Software Architect David Farber (Jun 24)
- Re: worth reading -- QOS Author is Motorola, Chief Software Architect David Farber (Jun 24)
- Re: worth reading -- QOS Author is Motorola, Chief Software Architect David Farber (Jun 24)
- Re: worth reading -- QOS Author is Motorola, Chief Software Architect David Farber (Jun 24)
- Re: worth reading -- QOS Author is Motorola, Chief Software Architect David Farber (Jun 26)