The MPLS WG Archive[Date Prev][Date Next][Thread Prev][Thread Next] [Date Index][Thread Index][Author Index][Subject Index] draft-ietf-mpls-diff-ext-07.txt
Francis, Regarding your example: To the best of my understanding AF11, AF12 and AF13 are three DSCPs belonging to the same PHB (AF1), expressing 3 different levels of drop precedence. So you have actually given an excellnet example of an L-LSP (with the scheduling behavior inferred from the label and drop precedence - from the EXP bits). A real example of an E-LSP would be an LSP carrying, say AF-1 and AF-2. Such a sharing carries with it some difficult questions, i.e.how to split the aggregate bandwidth allocated to an E-LSP between two (or more) PHBs with individual bandwidth reservations etc. ______________________________________________ With best regards, Sasha Vainshtein mailto: sasha@iprad.co.il phone: +972-3-7659993 (office) fax: +972-3-6487779 (office) >>-----Original Message----- >>From: owner-mpls@UU.NET [mailto:owner-mpls@UU.NET]On Behalf Of >>Francis Arts >>Sent: Thursday, November 30, 2000 9:39 AM >>To: mpls@UU.NET >>Subject: draft-ietf-mpls-diff-ext-07.txt >> >> >> >>Hello, >> >>The draft-ietf-mpls-diff-ext-07.txt defines 2 types of LSPs - E-LSPs and >>L-LSPs. For the DIFF SERV object defined in this draft, also 2 variants >>are defined -1 for E-LSPs and 1 for L-LSPs. >> >>I have the following question: Why have L-LSPs been defined (instead of >>defining just 1 type - E-LSPs)? According to my understanding all the >>capabilities of an L-LSP can be obtained by using an E-LSP with the >>proper PHB set. >> >>Example: Assume that we want to establish an L-LSP for AF1 traffic. We >>could achieve the same effect (an MPLS tunnel for Af1 traffic) by using >>an E-LSP with 3 exp <--> PHB mappings: >>** exp_af11 <--> AF11 >>** exp_af12 <--> AF12 >>** exp_af13 <--> AF13 >> >> >>Thanks for your answer. >> >>Kind regards, >> >> Francis. >> >>
|
|