Re: [PCN] Concensus questions from Thursday's PCN meeting
"Geib, Ruediger" <Ruediger.Geib@t-systems.com> Tue, 18 March 2008 10:37 UTC
Return-Path: <pcn-bounces@ietf.org>
X-Original-To: ietfarch-pcn-archive@core3.amsl.com
Delivered-To: ietfarch-pcn-archive@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 03BDB28C2C1; Tue, 18 Mar 2008 03:37:07 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -100.693
X-Spam-Level:
X-Spam-Status: No, score=-100.693 tagged_above=-999 required=5 tests=[AWL=-0.256, BAYES_00=-2.599, FH_RELAY_NODNS=1.451, HELO_MISMATCH_ORG=0.611, RDNS_NONE=0.1, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 7SJnxppWA5Vj; Tue, 18 Mar 2008 03:37:02 -0700 (PDT)
Received: from core3.amsl.com (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 7F86B28C475; Tue, 18 Mar 2008 03:37:02 -0700 (PDT)
X-Original-To: pcn@core3.amsl.com
Delivered-To: pcn@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 2248828C475 for <pcn@core3.amsl.com>; Tue, 18 Mar 2008 03:37:02 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 1ffRIeIQkmn1 for <pcn@core3.amsl.com>; Tue, 18 Mar 2008 03:36:59 -0700 (PDT)
Received: from tcmail12.telekom.de (tcmail12.telekom.de [217.5.214.82]) by core3.amsl.com (Postfix) with ESMTP id CF02E28C2C1 for <pcn@ietf.org>; Tue, 18 Mar 2008 03:36:58 -0700 (PDT)
Received: from s4de8psaans.mitte.t-com.de (s4de8psaans.mitte.t-com.de [10.151.180.168]) by tcmail11.telekom.de with ESMTP; Tue, 18 Mar 2008 11:34:40 +0100
Received: from S4DE8PSAANK.mitte.t-com.de ([10.151.229.10]) by s4de8psaans.mitte.t-com.de with Microsoft SMTPSVC(6.0.3790.3959); Tue, 18 Mar 2008 11:34:40 +0100
X-MimeOLE: Produced By Microsoft Exchange V6.5
Content-class: urn:content-classes:message
MIME-Version: 1.0
Date: Tue, 18 Mar 2008 11:36:02 +0100
Message-Id: <1B6169C658325341A3B8066E23919E1CF647C5@S4DE8PSAANK.mitte.t-com.de>
In-Reply-To: <000701c888e1$37a68c20$810c5982@dynamic.ewi.utwente.nl>
X-MS-Has-Attach:
X-MS-TNEF-Correlator:
Thread-Topic: [PCN] Concensus questions from Thursday's PCN meeting
Thread-Index: AciIBF++zKz17E/GQcib+W/rjB17BgABJ68gAABXALAAAifOMAAH2KywAACjCyAAAJeqgAAA4wtQAAGeymAAAM8HgAAA8S7AAAEXZnAAHn88IAAGS18AAADxzSA=
References: <BABC859E6D0B9A4D8448CC7F41CD2B0706181835@xmb-rtp-203.amer.cisco.com> <RrmbUrJK.1205679770.1867150.karagian@ewi.utwente.nl> <1B6169C658325341A3B8066E23919E1CF641B0@S4DE8PSAANK.mitte.t-com.de> <000001c88809$b2e73840$810c5982@dynamic.ewi.utwente.nl> <1B6169C658325341A3B8066E23919E1CF6423C@S4DE8PSAANK.mitte.t-com.de> <001301c88816$114dab60$810c5982@dynamic.ewi.utwente.nl> <1B6169C658325341A3B8066E23919E1CF644B1@S4DE8PSAANK.mitte.t-com.de> <000001c88835$998bcf60$810c5982@dynamic.ewi.utwente.nl> <1B6169C658325341A3B8066E23919E1CF6451A@S4DE8PSAANK.mitte.t-com.de> <000601c8883b$e3828950$810c5982@dynamic.ewi.utwente.nl> <1B6169C658325341A3B8066E23919E1CF64580@S4DE8PSAANK.mitte.t-com.de> <000901c88844$f35c1130$810c5982@dynamic.ewi.utwente.nl> <1B6169C658325341A3B8066E23919E1CF645A3@S4DE8PSAANK.mitte.t-com.de> <000a01c8884d$081c9790$810c5982@dynamic.ewi.utwente.nl> <1B6169C658325341A3B8066E23919E1CF64645@S4DE8PSAANK.mitte.t-com.de> <000701c888e1$37a68c20$810c5982@dynamic.ew i.utwent e.nl>
From: "Geib, Ruediger" <Ruediger.Geib@t-systems.com>
To: karagian@cs.utwente.nl
X-OriginalArrivalTime: 18 Mar 2008 10:34:40.0458 (UTC) FILETIME=[ABFE42A0:01C888E3]
Cc: pcn@ietf.org
Subject: Re: [PCN] Concensus questions from Thursday's PCN meeting
X-BeenThere: pcn@ietf.org
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: PCN WG list <pcn.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/pcn>, <mailto:pcn-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/pipermail/pcn>
List-Post: <mailto:pcn@ietf.org>
List-Help: <mailto:pcn-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/pcn>, <mailto:pcn-request@ietf.org?subject=subscribe>
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Sender: pcn-bounces@ietf.org
Errors-To: pcn-bounces@ietf.org
Hi Georgios see in line. Regards, Rudiger | -----Original Message----- | From: Georgios Karagiannis [mailto:karagian@cs.utwente.nl] | Sent: Tuesday, March 18, 2008 11:17 AM | To: Geib, Rüdiger | Cc: pcn@ietf.org | Subject: RE: [PCN] Concensus questions from Thursday's PCN meeting | | Hi Ruediger | | The problem that I have explained it does not have to do with | slow reaction time but it has to do with complete collapse of | the PCN domain operation. | This occurs when the admission control/flow termination | triggers at the egress are not activated when they should be | activated. | | | Here is a more concrete example: | | When ECMP routing is used, it means that flows that are | associated with the same ingress-egress-aggregate can follow | different paths. | Assume that there are 5 such paths that are carrying flows | associated with the same ingress-egress-aggregate. | | Consider now that one of the 5 paths fails and it is rerouted | to one of the other paths. | Assume that path1 is becoming congested and the other 3 | remaining paths are not congested. | I am also assuming in this case that the thresholds are well | configured. This is not realistic. | The egress will calculate CLE in the following way: | CLE = marked_1 /(unmarked_1 + unmarked_2 + unmarked_3 + unmarked_4 + | marked_1) | | I think that typcially the above calculated CLE will not | reach 1% when marked packets are preferentially dropped. | This problem is due to the fact that ECMP routing is used and | flows that are not participating at the congestion are | contributing to the calculation of the CLE. The flows that | are not passing through congested nodes will actually forward | unmarked packets to the egress node. | | When marked packets are preferentially dropped than marked_1 | can easily be smaller than 1% * (unmarked_1 + unmarked_2 + | unmarked_3 + unmarled_4 + marked_1). | In this case the admission control situation is not triggered | and a complete collapse of the PCN domain operation can occur! | | Thus: | If no ECMP routing would have been used than for the same | situation CLE = marked_1/unmarked_1 + marked_1), which can go | easily to 1%, if the C-A-R is well configured. | | But when ECMP routing is used and the above scenario is | applied then CLE = marked_1 /(unmarked_1 + unmarked_2 + | unmarked_3 + unmarled_4 + marked_1), which probaly will not | reach 1% when marked packets are preferentially dropped, | since marked_1 can be smaller than 1% * (unmarked_1 + | unmarked_2 + unmarked_3 + unmarled_4 + | marked_1) | | I hope that you now understand the problem! | | Best regards, | Georgios | | > -----Original Message----- | > From: Geib, Ruediger [mailto:Ruediger.Geib@t-systems.com] | > Sent: dinsdag 18 maart 2008 8:30 | > To: karagian@cs.utwente.nl | > Cc: pcn@ietf.org | > Subject: RE: [PCN] Concensus questions from Thursday's PCN meeting | > | > Hi Georgios, | > | > in the situation you describe, packet losses occur. This | will result | > in bad press, as the customers using PCN based services | were promised | > another type of service. | > | > In this situation it doesn't matter whether or not ECMP is deployed | > and it also doesn't matter whether termination is fair or not. The | > important event is: packet losses occur (in one of your examples | > several routers drop packets). The drops are the only | relevant issue. | > Whether service resumes after 5 seconds due to extremly well | > engineered termination or after 10 seconds after a | sufficient number | > of customers hang up is not important. | > I can't recall having read anytime in the news "Major | network outage - | > but termination was fair." I can only recall having seen the first | > part. | > | > I'm sure you're happy in adapting your example, as you do all the | > time. I'm having work to do, but maybe someone else is | interested in | > continuing discusion. I think, I've made my point. | > | > Regards, | > | > Rudiger | > | > | -----Original Message----- | > | From: Georgios Karagiannis [mailto:karagian@cs.utwente.nl] | > | Sent: Monday, March 17, 2008 5:36 PM | > | To: Geib, Rüdiger | > | Cc: pcn@ietf.org | > | Subject: RE: [PCN] Concensus questions from Thursday's PCN meeting | > | | > | Hi Ruediger | > | | > | Please note that what I have described holds also for the | situation | > | that the routers are well configured, but a catastrophic | > event occur | > | and at the same time ECMP routing is used. | > | | > | Best regards, | > | Georgios | > | | > | | > | > -----Original Message----- | > | > From: Geib, Ruediger [mailto:Ruediger.Geib@t-systems.com] | > | > Sent: maandag 17 maart 2008 17:10 | > | > To: karagian@cs.utwente.nl | > | > Cc: pcn@ietf.org | > | > Subject: RE: [PCN] Concensus questions from Thursday's | PCN meeting | > | > | > | > Hi Georgios, | > | > | > | > what I say is if PCN decides to deal with catastrophic | events in | > | > combination with misconfigured routers, then it will end up | > | > standardising solutions for academic problems. | > | > | > | > Regards, | > | > | > | > Rudiger | > | > | > | > | > | > | -----Original Message----- | > | > | From: Georgios Karagiannis [mailto:karagian@cs.utwente.nl] | > | > | Sent: Monday, March 17, 2008 4:38 PM | > | > | To: Geib, Rüdiger | > | > | Cc: pcn@ietf.org | > | > | Subject: RE: [PCN] Concensus questions from Thursday's | > PCN meeting | > | > | | > | > | Hi Ruediger | > | > | | > | > | I do not understand what you try to say. Please note | that flow | > | > | termination is used for the below explained reason, see | > | below text | > | > | taken from the PCN architecture draft: | > | > | | > | > | o The termination mechanism complements admission | control. It | > | > | allows the network to recover from sudden | > | unexpected surges of | > | > | PCN-traffic on some links, thus restoring QoS to | > | the remaining | > | > | flows. Such scenarios are expected to be rare but not | > | > | impossible. | > | > | They can be caused by large network failures that | > | > redirect lots | > | > | of | > | > | admitted PCN-traffic to other links, or by | > | malfunction of the | > | > | measurement-based admission control in the presence | > | > of admitted | > | > | flows that send for a while with an atypically low | > | > rate and then | > | > | increase their rates in a correlated way. | > | > | | > | > | Best regards, | > | > | Georgios | > | > | | > | > | | > | > | | > | > | > -----Original Message----- | > | > | > From: Geib, Ruediger [mailto:Ruediger.Geib@t-systems.com] | > | > | > Sent: maandag 17 maart 2008 16:32 | > | > | > To: karagian@cs.utwente.nl | > | > | > Cc: pcn@ietf.org | > | > | > Subject: RE: [PCN] Concensus questions from Thursday's | > | PCN meeting | > | > | > | > | > | > Hi Georgios, | > | > | > | > | > | > QoS on redundant links means, that one link is able to | > | > | carry the QoS | > | > | > traffic of the other. This may not hold for non QoS | > | > | traffic. This is, | > | > | > what typical carrier backbones providing QoS are | > engineered for | > | > | > (commonly known as traffic engineering). | > | > | > | > | > | > Should a misconfiguration occur, things may go wrong. | > | > | > Operational staff corrects this a soon as the | > | misconfiguration is | > | > | > identified. | > | > | > | > | > | > The engineering of carrier backbones is often done with | > | automated | > | > | > tools, leaving little room for human errors. | > | > | > However maintenance work may result in humanne system | > | > | interference. I | > | > | > doubt that Telekom's engineering department is willing to | > | > | accept extra | > | > | > protocol and router complexitiy to deal with operator | > | > configuration | > | > | > errors. While I time and again hear of the worst | > | > | misconfigs, I can't | > | > | > recall that my colleagues try to adapt standardisation | > | afterwards. | > | > | > | > | > | > Regards, | > | > | > | > | > | > Rudiger | > | > | > | > | > | > | -----Original Message----- | > | > | > | From: Georgios Karagiannis [mailto:karagian@cs.utwente.nl] | > | > | > | Sent: Monday, March 17, 2008 3:34 PM | > | > | > | To: Geib, Rüdiger | > | > | > | Cc: pcn@ietf.org | > | > | > | Subject: RE: [PCN] Concensus questions from Thursday's | > | > PCN meeting | > | > | > | | > | > | > | Hi Ruediger | > | > | > | | > | > | > | Please note that typical IP networks (thus not | > | MPLS/GMPLS based | > | > | > | domains) can recover from failures by using rerouting. | > | > | > | Or do you know another standardized mechanism in | typical IP | > | > | > networks | > | > | > | (thus not MPLS/GMPLS based) that can provide another kind | > | > | > of recovery | > | > | > | from failures. | > | > | > | | > | > | > | Furthermore, I will just explain the example again | > | with maximum | > | > | > | capacity of | > | > | > | pathh2 being equal | > | > | > | to maximum capacity of path1 = C. | > | > | > | Please explain what you do not understand in this example! | > | > | > | | > | > | > | Here are the assumptions: | > | > | > | * In the PCN domain we assume that ECMP routing | is possible! | > | > | > | * ingress-eggress-aggregate can contain flows that are | > | > | > passing through | > | > | > | one path or more paths (when ECMP routing is used). | > | > | > | | > | > | > | * routers are currently dropping packets randomly. Thus | > | > | marked and | > | > | > | unmarked packets will be | > | > | > | dropped randomly | > | > | > | | > | > | > | * When excess rate measurements are used, the | triggering of | > | > | > admission | > | > | > | control and flow | > | > | > | termination are done at the egress by using the CLE. One | > | > | > example of | > | > | > | this trigger is: | > | > | > | CLE > 1%. Note that CLE = marked/(unmarked + marked). | > | > | This means | > | > | > | that if this trigger is not activated while a severe | > | > | > | congestion occurs in the PCN domain, then the operation | > | > | > of the PCN | > | > | > | domain will completely collapse. | > | > | > | | > | > | > | | > | > | > | My statement is: | > | > | > | If the routers do not preferentially drop marked packets | > | > | > then the PCN | > | > | > | domain operation, even in corner case and misconfiguration | > | > | > situations, | > | > | > | is more robust and more stable than in the situation that | > | > | > the router | > | > | > | is preferentially dropping marked packets. | > | > | > | | > | > | > | The below example show such a corner case! | > | > | > | | > | > | > | | > | > | > | I will describe two situations: | > | > | > | | > | > | > | Situation 1: | > | > | > | Consider that an ingress-egress-aggregate due to ECMP | > | > routing it | > | > | > | includes flows that are passing from at least two paths. | > | > | > | Assume that path1 supports a maximum bandwidth | > capacity of C. | > | > | > | Now consider that the maximum bandwdith capacity of | > | path2 is C. | > | > | > | Consider also that both paths are fully utilized. | > | > | > | | > | > | > | Consider that preferentially marked packets are | dropped and | > | > | > that path2 | > | > | > | fails. | > | > | > | Now assume that all (maximum) traffic passing through | > | > | path2 will be | > | > | > | rerouted through path1. | > | > | > | If a bottleneck router located in path1, say Rbott, is | > | > | > misconfigured | > | > | > | (i.e., from the point of view of having wrongly too high | > | > | > | configured-admissible-rate value), or even if it is well | > | > | > configured, | > | > | > | then it can be possible that CLE measured at the egress | > | > | > cannot reach | > | > | > | 1%. | > | > | > | Note that | > | > | > | CLE = marked packets/ (marked packets + unmarked packets). | > | > | > | This is because Rbott will just allow an excess rate to | > | > | > pass through | > | > | > | that is | > | > | > | | > | > | > | equal to: C - configured-admissible-rate. The rest of | > | > the marked | > | > | > | packets, so the rest of the excess rate, will be dropped | > | > | by Rbott, | > | > | > | since this router is preferentially dropping | marked packets. | > | > | > | Furthermore, note that due to the ECMP routing | > flows that are | > | > | > | belonging to the same ingress-egress-aggregate and that | > | > | are passing | > | > | > | through another path, | > | > | > | | > | > | > | than path1, which is not congested, will forward traffic | > | > | > towards the | > | > | > | egress node that will be unmarked. | > | > | > | This will mean that the CLE > 1% will not be triggered, | > | > | > this will mean | > | > | > | that the flow termination will not be triggered and that | > | > | > the operation | > | > | > | of the PCN domain will collapse completely. | > | > | > | | > | > | > | | > | > | > | Situation 2: | > | > | > | Now consider that the router just operates as | > | > currently, i.e., no | > | > | > | preferential drop, randomly dropping marked and | > | > unmarked packets. | > | > | > | Consider also that the two paths described in Situation 1 | > | > | above are | > | > | > | used, ECMP routing is used, and that they are fully | > utilized. | > | > | > | Assume also that path2 fails and that the path2 traffic is | > | > | > rerouted on | > | > | > | path1. | > | > | > | Now the CLE value will have a higher probability of | > | > reaching the | > | > | > | triggering value of 1%. | > | > | > | | > | > | > | This is because the routers in path1 will drop randomly | > | > | marked and | > | > | > | unmarked packets. | > | > | > | It is more certain that in this case the CLE will reach 1% | > | > | > due to the | > | > | > | following reason. | > | > | > | In this example it is assumed that the maximum | > | > bandwidth capacity | > | > | > | supported by path2 is C. | > | > | > | This means that after rerouting the traffic from | path2 into | > | > | > path1, the | > | > | > | ratio between marked packets and unmarked packets that | > | > | are passing | > | > | > | thorugh path1 can be equal or higher than 100%. | > | > | > | Note that the routers in path1 will mark the excess rate | > | > | > above C, thus | > | > | > | 1*C (i.e., the rerouted traffic from path2) as marked. | > | > | > | The above observation holds also for the situation that | > | > | the routers | > | > | > | are preferentially dropping unmarked packets. | > | > | > | | > | > | > | Thus when routers are preferentially dropping marked | > | > packets the | > | > | > | robustness of the PCN domain operation is | decreasing and in | > | > | > some cases | > | > | > | it severely decreases, which could cause the complete | > | > | > collapse of the | > | > | > | PCN domain operation. | > | > | > | | > | > | > | Best regards, | > | > | > | Georgios | > | > | > | | > | > | > | | > | > | > | | > | > | > | | > | > | > | > -----Original Message----- | > | > | > | > From: Geib, Ruediger | [mailto:Ruediger.Geib@t-systems.com] | > | > | > | > Sent: maandag 17 maart 2008 15:26 | > | > | > | > To: karagian@cs.utwente.nl | > | > | > | > Cc: pcn@ietf.org | > | > | > | > Subject: RE: [PCN] Concensus questions from Thursday's | > | > | PCN meeting | > | > | > | > | > | > | > | > Hi Georgios, | > | > | > | > | > | > | > | > Unfortunately you missed my point. | > | > | > | > | > | > | > | > No such thing as you describe has been, is or | will for the | > | > | > | foreseeable | > | > | > | > future be engineered in the IP backbone of the company | > | > | > I'm working | > | > | > | > for. | > | > | > | > | > | > | > | > My expectation on the backbone of any carrier offering | > | > | > | transport QoS | > | > | > | > is, that it is engineered to survive probable failures | > | > | > (like single | > | > | > | > link losses) without impact on the performance of QoS | > | > | > | traffic. This is | > | > | > | > where we should start from, if we discuss PCN. | From there | > | > | > | on, PCN is | > | > | > | > able to add value. | > | > | > | > | > | > | > | > Links without traffic engineering followed by | > | > | > misconfigured routers | > | > | > | > aren't the arguments which will convince either product | > | > | > | management or | > | > | > | > backbone engineering of the company I work for to | > | deploy PCN. | > | > | > | > | > | > | > | > As a general comment: I don't regard PCN as a plug and | > | > | > play toy for | > | > | > | > the operational equivalent of a 'script | kiddie'. If others | > | > | > | share that | > | > | > | > view and our documents don't state that, we | > should add text. | > | > | > | > | > | > | > | > Regards, | > | > | > | > | > | > | > | > Rudiger | > | > | > | > | > | > | > | > | -----Original Message----- | > | > | > | > | From: Georgios Karagiannis | > [mailto:karagian@cs.utwente.nl] | > | > | > | > | Sent: Monday, March 17, 2008 2:49 PM | > | > | > | > | To: Geib, Rüdiger | > | > | > | > | Cc: pcn@ietf.org | > | > | > | > | Subject: RE: [PCN] Concensus questions from Thursday's | > | > | > PCN meeting | > | > | > | > | | > | > | > | > | Hi Ruediger | > | > | > | > | | > | > | > | > | You missed my point! | > | > | > | > | The explanation applies to all types of IP networks. | > | > | > | > | Please note that the problem occurs also when | path2 and | > | > | > path1 use | > | > | > | > | exactly the same maximum capacity, say C! | > | > | > | > | | > | > | > | > | In this case, if pathh2 fails then path1 will see | > | an excess | > | > | > | > rate equal | > | > | > | > | to at least C. | > | > | > | > | The rest of the explanation is identical to | the one used | > | > | > | during my | > | > | > | > | previous example! | > | > | > | > | | > | > | > | > | | > | > | > | > | | > | > | > | > | Best regards, | > | > | > | > | Georgios | > | > | > | > | | > | > | > | > | | > | > | > | > | | > | > | > | > | | > | > | > | > | | > | > | > | > | > -----Original Message----- | > | > | > | > | > From: Geib, Ruediger | > | [mailto:Ruediger.Geib@t-systems.com] | > | > | > | > | > Sent: maandag 17 maart 2008 14:34 | > | > | > | > | > To: karagian@cs.utwente.nl | > | > | > | > | > Cc: pcn@ietf.org | > | > | > | > | > Subject: RE: [PCN] Concensus questions from | Thursday's | > | > | > | PCN meeting | > | > | > | > | > | > | > | > | > | > Hi Georgios, | > | > | > | > | > | > | > | > | > | > What you describe isn't backbone traffic | engineering. | > | > | > What you | > | > | > | > | > describe may happen in corporate VPNs, where a | > | DSL access | > | > | > | > | is used to | > | > | > | > | > back up a WAN Fast Ethernet access or the | > like. I don't | > | > | > | > | object to have | > | > | > | > | > standards on PCN for VPNs, but I'm not | > | interested in this | > | > | > | > issue for | > | > | > | > | > now. | > | > | > | > | > | > | > | > | > | > If you know any carrier who's operating his | > | network with a | > | > | > | > | > 16:1 load balancing for QoS traffic on any | > | particular link | > | > | > | > | set, please | > | > | > | > | > publish the name, we are all interested in | hearing it. | > | > | > | > | > | > | > | > | > | > Regards, | > | > | > | > | > | > | > | > | > | > Rudiger | > | > | > | > | > | > | > | > | > | > | > | > | > | > | > | > | > | > | > | > | -----Original Message----- | > | > | > | > | > | From: Georgios Karagiannis | > | > [mailto:karagian@cs.utwente.nl] | > | > | > | > | > | Sent: Monday, March 17, 2008 11:03 AM | > | > | > | > | > | To: Geib, Rüdiger | > | > | > | > | > | Cc: pcn@ietf.org | > | > | > | > | > | Subject: RE: [PCN] Concensus questions from | > Thursday's | > | > | > | > PCN meeting | > | > | > | > | > | | > | > | > | > | > | Hi Ruediger | > | > | > | > | > | | > | > | > | > | > | I will try to explain this more clearly. | > | > | > | > | > | Here are the assumptions: | > | > | > | > | > | * In the PCN domain we assume that ECMP routing | > | > | is possible! | > | > | > | > | > | * ingress-eggress-aggregate can contain | > flows that are | > | > | > | > | > passing through | > | > | > | > | > | one path or more paths (when ECMP routing | is used). | > | > | > | > | > | | > | > | > | > | > | * routers are currently dropping packets | > | randomly. Thus | > | > | > | > | marked and | > | > | > | > | > | unmarked packets will be | > | > | > | > | > | dropped randomly | > | > | > | > | > | | > | > | > | > | > | * When excess rate measurements are used, the | > | > | triggering of | > | > | > | > | > admission | > | > | > | > | > | control and flow | > | > | > | > | > | termination are done at the egress by using | > | > the CLE. One | > | > | > | > | > example of | > | > | > | > | > | this trigger is: | > | > | > | > | > | CLE > 1%. Note that CLE = marked/(unmarked | > | + marked). | > | > | > | > | This means | > | > | > | > | > | that if this trigger is not activated | while a severe | > | > | > | > | > | congestion occurs in the PCN domain, then the | > | > operation | > | > | > | > | > of the PCN | > | > | > | > | > | domain will completely collapse. | > | > | > | > | > | | > | > | > | > | > | | > | > | > | > | > | My statement is: | > | > | > | > | > | If the routers do not preferentially drop | > | marked packets | > | > | > | > | > then the PCN | > | > | > | > | > | domain operation, even in corner case and | > | > misconfiguration | > | > | > | > | > situations, | > | > | > | > | > | is more robust and more stable than in the | > | > situation that | > | > | > | > | > the router | > | > | > | > | > | is preferentially dropping marked packets. | > | > | > | > | > | | > | > | > | > | > | The below example show such a corner case! | > | > | > | > | > | | > | > | > | > | > | | > | > | > | > | > | I will describe two situations: | > | > | > | > | > | | > | > | > | > | > | Situation 1: | > | > | > | > | > | Consider that an ingress-egress-aggregate | > due to ECMP | > | > | > | > routing it | > | > | > | > | > | includes flows that are passing from at least | > | two paths. | > | > | > | > | > | Assume that path1 supports a maximum bandwidth | > | > | > capacity of C. | > | > | > | > | > | Now consider | > | > | > | > | > | that the maximum bandwdith capacity of | > path2 is 15*C. | > | > | > | > | > | Consider also that both paths are fully utilized. | > | > | > | > | > | | > | > | > | > | > | Consider that preferentially marked packets are | > | > | dropped and | > | > | > | > | > that path2 | > | > | > | > | > | fails. | > | > | > | > | > | Now assume that all (maximum) traffic | > passing through | > | > | > | > | path2 will be | > | > | > | > | > | rerouted through path1. | > | > | > | > | > | If a bottleneck router located in path1, | > say Rbott, is | > | > | > | > | > misconfigured | > | > | > | > | > | (i.e., from the point of view of having wrongly | > | > too high | > | > | > | > | > | configured-admissible-rate value), or even if | > | it is well | > | > | > | > | > configured, | > | > | > | > | > | then it can be possible that CLE measured at | > | the egress | > | > | > | > | > cannot reach | > | > | > | > | > | 1%. | > | > | > | > | > | Note that | > | > | > | > | > | CLE = marked packets/ (marked packets + | > | > unmarked packets). | > | > | > | > | > | This is because Rbott will just allow an | > | excess rate to | > | > | > | > | > pass through | > | > | > | > | > | that is | > | > | > | > | > | | > | > | > | > | > | equal to: C - configured-admissible-rate. | > The rest of | > | > | > | > the marked | > | > | > | > | > | packets, so the rest of the excess rate, will | > | be dropped | > | > | > | > | by Rbott, | > | > | > | > | > | since this router is preferentially dropping | > | > | marked packets. | > | > | > | > | > | Furthermore, note that due to the ECMP routing | > | > | > flows that are | > | > | > | > | > | belonging to the same | > | ingress-egress-aggregate and that | > | > | > | > | are passing | > | > | > | > | > | through another path, | > | > | > | > | > | | > | > | > | > | > | than path1, which is not congested, will | > | forward traffic | > | > | > | > | > towards the | > | > | > | > | > | egress node that will be unmarked. | > | > | > | > | > | This will mean that the CLE > 1% will not be | > | triggered, | > | > | > | > | > this will mean | > | > | > | > | > | that the flow termination will not be | > | triggered and that | > | > | > | > | > the operation | > | > | > | > | > | of the PCN domain will collapse completely. | > | > | > | > | > | | > | > | > | > | > | | > | > | > | > | > | Situation 2: | > | > | > | > | > | Now consider that the router just operates as | > | > | > | > currently, i.e., no | > | > | > | > | > | preferential drop, randomly dropping marked and | > | > | > | > unmarked packets. | > | > | > | > | > | Consider also that the two paths described in | > | > Situation 1 | > | > | > | > | above are | > | > | > | > | > | used, ECMP routing is used, and that they | are fully | > | > | > utilized. | > | > | > | > | > | Assume also that path2 fails and that the path2 | > | > traffic is | > | > | > | > | > rerouted on | > | > | > | > | > | path1. | > | > | > | > | > | Now the CLE value will have a higher | probability of | > | > | > | > reaching the | > | > | > | > | > | triggering value of 1%. | > | > | > | > | > | | > | > | > | > | > | This is because the routers in path1 will | > | drop randomly | > | > | > | > | marked and | > | > | > | > | > | unmarked packets. | > | > | > | > | > | It is more certain that in this case the CLE | > | > will reach 1% | > | > | > | > | > due to the | > | > | > | > | > | following reason. | > | > | > | > | > | In this example it is assumed that the maximum | > | > | > | > bandwidth capacity | > | > | > | > | > | supported by path2 is 15*C. | > | > | > | > | > | This means that after rerouting the traffic from | > | > | path2 into | > | > | > | > | > path1, the | > | > | > | > | > | ratio between marked packets and unmarked | > packets that | > | > | > | > | are passing | > | > | > | > | > | thorugh path1 can be equal or higher than 16. | > | > | > | > | > | Note that the routers in path1 will mark the | > | excess rate | > | > | > | > | > above C, thus | > | > | > | > | > | 16 * C (i.e., the rerouted traffic from path2) | > | > as marked. | > | > | > | > | > | The above observation holds also for the | > | situation that | > | > | > | > | the routers | > | > | > | > | > | are preferentially dropping unmarked packets. | > | > | > | > | > | | > | > | > | > | > | Thus when routers are preferentially | dropping marked | > | > | > | > packets the | > | > | > | > | > | robustness of the PCN domain operation is | > | > | decreasing and in | > | > | > | > | > some cases | > | > | > | > | > | it severely decreases, which could cause | > the complete | > | > | > | > | > collapse of the | > | > | > | > | > | PCN domain operation. | > | > | > | > | > | | > | > | > | > | > | Best regards, | > | > | > | > | > | Georgios | > | > | > | > | > | | > | > | > | > | > | | > | > | > | > | > | | > | > | > | > | > | | > | > | > | > | > | | > | > | > | > | > | > -----Original Message----- | > | > | > | > | > | > From: Geib, Ruediger | > | > | [mailto:Ruediger.Geib@t-systems.com] | > | > | > | > | > | > Sent: maandag 17 maart 2008 9:42 | > | > | > | > | > | > To: karagian@cs.utwente.nl | > | > | > | > | > | > Cc: pcn@ietf.org | > | > | > | > | > | > Subject: RE: [PCN] Concensus questions from | > | Thursday's | > | > | > | > | PCN meeting | > | > | > | > | > | > | > | > | > | > | > | > Hi Georgios, | > | > | > | > | > | > | > | > | > | > | > | > PCN is designed for deployment in traffic | > engineered | > | > | > | > networks. | > | > | > | > | > | > Please decribe, how to engineer a high | performance | > | > | > | > | > carrier backbone | > | > | > | > | > | > and then review the assumptions your discussion | > | > | > is based on. | > | > | > | > | > | > | > | > | > | > | > | > Regards, | > | > | > | > | > | > | > | > | > | > | > | > Rudiger | > | > | > | > | > | > | > | > | > | > | > | > | -----Original Message----- | > | > | > | > | > | > | From: Georgios Karagiannis | > | > | > [mailto:karagian@cs.utwente.nl] | > | > | > | > | > | > | Sent: Monday, March 17, 2008 9:34 AM | > | > | > | > | > | > | To: Geib, Rüdiger | > | > | > | > | > | > | Cc: pcn@ietf.org | > | > | > | > | > | > | Subject: RE: [PCN] Concensus questions from | > | > Thursday's | > | > | > | > | > PCN meeting | > | > | > | > | > | > | | > | > | > | > | > | > | Hi Ruediger | > | > | > | > | > | > | | > | > | > | > | > | > | What do you mean? | > | > | > | > | > | > | Do you mean that you do not want to discuss | > | > | corner cases | > | > | > | > | > | > (ECMP related | > | > | > | > | > | > | cases) that could collapse the PCN domain | > | > | > | > | > | > | operation? | > | > | > | > | > | > | What I am saying is that if the routers do not | > | > | > | > | > | preferentially drop | > | > | > | > | > | > | marked packets then the PCN domain | operation is | > | > | > | more robust | > | > | > | > | > | > and more | > | > | > | > | > | > | stable than in the situation that the | router is | > | > | > | > | preferentially | > | > | > | > | > | > | dropping marked packets. | > | > | > | > | > | > | | > | > | > | > | > | > | Are you saying that this statement is | not right? | > | > | > | > | > | > | | > | > | > | > | > | > | | > | > | > | > | > | > | Best regards, | > | > | > | > | > | > | Georgios | > | > | > | > | > | > | | > | > | > | > | > | > | > | > | > | > | > | | > | > | > | > | > | | > | > | > | > | > | | > | > | > | > | > | > | > | > | > | | > | > | > | > | | > | > | > | > | | > | > | > | > | > | > | > | | > | > | > | | > | > | > | | > | > | > | > | > | | > | > | | > | > | | > | > | > | | > | | > | | > | | | _______________________________________________ PCN mailing list PCN@ietf.org https://www.ietf.org/mailman/listinfo/pcn
- [PCN] Concensus questions from Thursday's PCN mee… Steven Blake
- Re: [PCN] Concensus questions from Thursday's PCN… toby.moncaster
- Re: [PCN] Concensus questions from Thursday's PCN… Anna Charny (acharny)
- Re: [PCN] Concensus questions from Thursday's PCN… Steven Blake
- Re: [PCN] Concensus questions from Thursday's PCN… Georgios Karagiannis
- Re: [PCN] Concensus questions from Thursday's PCN… Anna Charny (acharny)
- Re: [PCN] Concensus questions from Thursday's PCN… Anna Charny (acharny)
- [PCN] [Fwd: RE: Concensus questions from Thursday… Steven Blake
- Re: [PCN] Concensus questions from Thursday's PCN… Georgios Karagiannis
- Re: [PCN] Concensus questions from Thursday's PCN… Georgios Karagiannis
- Re: [PCN] Concensus questions from Thursday's PCN… Geib, Ruediger
- Re: [PCN] Concensus questions from Thursday's PCN… Georgios Karagiannis
- Re: [PCN] Concensus questions from Thursday's PCN… Geib, Ruediger
- Re: [PCN] Concensus questions from Thursday's PCN… Georgios Karagiannis
- Re: [PCN] Concensus questions from Thursday's PCN… Geib, Ruediger
- Re: [PCN] Concensus questions from Thursday's PCN… Georgios Karagiannis
- Re: [PCN] Concensus questions from Thursday's PCN… Geib, Ruediger
- Re: [PCN] Concensus questions from Thursday's PCN… Georgios Karagiannis
- Re: [PCN] Concensus questions from Thursday's PCN… Geib, Ruediger
- Re: [PCN] Concensus questions from Thursday's PCN… Georgios Karagiannis
- Re: [PCN] Concensus questions from Thursday's PCN… Geib, Ruediger
- Re: [PCN] Concensus questions from Thursday's PCN… Wei Gengyu
- Re: [PCN] Concensus questions from Thursday's PCN… Georgios Karagiannis
- Re: [PCN] Concensus questions from Thursday's PCN… Georgios Karagiannis
- Re: [PCN] Concensus questions from Thursday's PCN… Geib, Ruediger
- Re: [PCN] Concensus questions from Thursday's PCN… Georgios Karagiannis
- Re: [PCN] Concensus questions from Thursday's PCN… Geib, Ruediger
- Re: [PCN] Concensus questions from Thursday's PCN… Georgios Karagiannis
- Re: [PCN] Concensus questions from Thursday's PCN… Geib, Ruediger
- Re: [PCN] Concensus questions from Thursday's PCN… Steven Blake
- Re: [PCN] Concensus questions from Thursday's PCN… Geib, Ruediger
- Re: [PCN] Concensus questions from Thursday's PCN… Georgios Karagiannis
- Re: [PCN] Concensus questions from Thursday's PCN… Geib, Ruediger
- Re: [PCN] Concensus questions from Thursday's PCN… Steven Blake
- Re: [PCN] Concensus questions from Thursday's PCN… Georgios Karagiannis
- Re: [PCN] Concensus questions from Thursday's PCN… Geib, Ruediger
- Re: [PCN] Concensus questions from Thursday's PCN… Georgios Karagiannis
- Re: [PCN] Concensus questions from Thursday's PCN… Geib, Ruediger
- Re: [PCN] Concensus questions from Thursday's PCN… philip.eardley
- Re: [PCN] Concensus questions from Thursday's PCN… Michael Menth
- Re: [PCN] Concensus questions from Thursday's PCN… Georgios Karagiannis
- Re: [PCN] Concensus questions from Thursday's PCN… Steven Blake
- Re: [PCN] Concensus questions from Thursday's PCN… Georgios Karagiannis
- Re: [PCN] Concensus questions from Thursday's PCN… philip.eardley
- Re: [PCN] Concensus questions from Thursday's PCN… Georgios Karagiannis
- Re: [PCN] Concensus questions from Thursday's PCN… philip.eardley
- Re: [PCN] Concensus questions from Thursday's PCN… toby.moncaster
- Re: [PCN] Concensus questions from Thursday's PCN… Anna Charny (acharny)
- Re: [PCN] Concensus questions from Thursday's PCN… Anna Charny (acharny)
- Re: [PCN] Concensus questions from Thursday's PCN… Wei Gengyu
- Re: [PCN] Concensus questions from Thursday's PCN… Georgios Karagiannis
- Re: [PCN] Concensus questions from Thursday's PCN… Georgios Karagiannis
- Re: [PCN] Concensus questions from Thursday's PCN… philip.eardley
- Re: [PCN] Concensus questions from Thursday's PCN… Georgios Karagiannis
- Re: [PCN] Concensus questions from Thursday's PCN… Geib, Ruediger
- Re: [PCN] Concensus questions from Thursday's PCN… Anna Charny (acharny)
- Re: [PCN] Concensus questions from Thursday's PCN… Anna Charny (acharny)
- [PCN] Fw: Concensus questions from Thursday's PCN… Wei Gengyu
- [PCN] On pcn and overloads (was: Concensus questi… Anna Charny (acharny)
- Re: [PCN] On pcn and overloads (was: Concensus qu… Geib, Ruediger
- Re: [PCN] On pcn and overloads (was: Concensus qu… Anna Charny (acharny)
- Re: [PCN] On pcn and overloads (was: Concensus qu… toby.moncaster
- Re: [PCN] Concensus questions from Thursday's PCN… Georgios Karagiannis
- Re: [PCN] Concensus questions from Thursday's PCN… Georgios Karagiannis
- [PCN] Georgios's example philip.eardley
- Re: [PCN] Concensus questions from Thursday's PCN… Steven Blake
- Re: [PCN] Concensus questions from Thursday's PCN… Georgios Karagiannis
- Re: [PCN] Concensus questions from Thursday's PCN… Georgios Karagiannis
- Re: [PCN] Concensus questions from Thursday's PCN… Michael Menth
- Re: [PCN] Concensus questions from Thursday's PCN… Michael Menth
- Re: [PCN] Concensus questions from Thursday's PCN… philip.eardley