RFC 2379
RSVP over ATM Implementation Guidelines
RFC 2379: RSVP over ATM Implementation Guidelines 🎀
- Written by L. Berger (that's Lou Berger 💅) from FORE Systems
- Dated August 1998 🗓️
- Category: Best Current Practice (it's BCP 24, so it's basically the official "do it this way" guide ✨)
- Part of the Network Working Group crew 🧚
- The title block has no Obsoletes or Updates entry, so it doesn't list any documents it replaces or changes 🦋
Status of this Memo 📜
Okay so this one is an Internet Best Current Practice for the whole Internet Community 🌐 It's also asking everyone for discussion and suggestions for improvements, so bring your ideas 💖 Distribution of this memo is unlimited, which means share it with everyone 🫶
Copyright Notice ©️
Copyright (C) The Internet Society (1998). All Rights Reserved. 🔒
Abstract ✨
This memo gives specific implementation guidelines for running RSVP over ATM switched virtual circuits (SVCs) 🦋 The general problem is discussed in [6]. Implementation requirements are discussed in [2]. Integrated Services to ATM service mappings are covered in [3]. Put all of those together and you get the full background and info you need to implement Integrated Services and RSVP over ATM 💅
1. Introduction 🌸
This memo is about running IP over ATM in an environment where SVCs carry QoS flows and RSVP acts as the internet level QoS signaling protocol 🔥 It applies when you're using CLIP/ION, LANE2.0, and MPOA methods for IP over ATM. The general issues with running RSVP[4] over ATM have been covered in a bunch of papers, including [6] and some earlier work 📚 This document is meant to be a companion to [6,2] and a guide for implementers, so you should be familiar with both of those docs first, no cap 🙈
This doc gives a recommended set of functionality for implementations using ATM UNI3.x and 4.0, while still leaving room for more sophisticated approaches 💜 We expect some vendors to also offer some of the fancier approaches described in [6], and some networks to only use those approaches. The recommended set is defined so you get predictability and interoperability between different implementations, which is honestly the whole point 👑 Requirements for RSVP over ATM implementations are in [2] 🎀
This document uses the same terms and assumptions stated in [2]. Also, the key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in RFC 2119 [5] 🧠
2. Implementation Recommendations 🍓
This section gives implementation guidelines for RSVP over ATM 🦋 A few recommendations apply to all RSVP sessions, both unicast and multicast. There are also recommendations that are unique to each session type 💖
2.1 RSVP Message VC Usage 📨
The general issues about which VC should carry RSVP messages are covered in [6]. It talks about several implementation options, including: mixed control and data, single control VC per session, single control VC multiplexed among sessions, and multiple VCs multiplexed among sessions 🌈 QoS for control VCs is also discussed there. We're not repeating the general discussion here, so go check [6] for the detailed info 🫶
RSVP over ATM implementations SHOULD send RSVP control (messages) over the best effort data path, see figure 1 ✨ It's permissible to let a user override this behavior 💅 This approach keeps VC requirements low, because the best effort data path has to exist anyway for RSVP sessions to be established and for RSVP reservations to be initiated 🧚 The specific best effort paths RSVP uses are: for unicast, the same VC used to reach the unicast destination; and for multicast, the same VC used for best effort traffic destined to the IP multicast group. For multicast there may be another best effort VC that carries session data traffic, meaning data that is both in the multicast group and matching a session's protocol and port 👑
Data Flow ==========>
QoS VCs
+-----+ --------------> +----+
| | --------------> | |
| Src | | R1 |
| | Best Effort VC(s) | |
+-----+ <-----------------> +----+
/\
||
||
RSVP Control
Messages
Figure 1: RSVP Control Message VC UsageThe downside is that best effort VCs may not provide the reliability RSVP needs 🥺 However, the best effort path is expected to meet RSVP reliability requirements in most networks. That's especially true since RSVP can handle a certain amount of packet loss without losing state synchronization 🦋
2.2 Aggregation 🧃
As [6] discusses, data for multiple RSVP sessions can be sent over the same shared VCs. Implementing these "aggregation" models is still a matter for research 🔬 So for now, RSVP over ATM implementations SHOULD use independent VCs for each RSVP reservation 💅
2.3 Short-Cuts ⚡
Short-cuts let ATM attached routers and hosts directly set up point-to-point VCs across LIS boundaries, meaning the VC endpoints are on different IP subnets 🌍 Short-cut support for unicast traffic is defined in [7] and [1]. People have asked whether short-cuts and RSVP can work together, and the main worry is handling asymmetric short-cuts 🤔 The specific question is how RSVP copes when a downstream short-cut doesn't have a matching upstream short-cut. In that case, shown in figure 2, PATH and RESV messages follow different paths 😭
______
/ \
+-------- / Router \ <-------+
| \ / | <....... RESVs Follow
| \______/ | Hop-by-hop Path
| |
| |
V QoS VCs |
+-----+ ==============> +----+
| | ==============> | |
| Src | | R1 |
| | Best Effort VC(s) | |
+-----+ <=================> +----+
/\
:: Data Paths:
:: ----> Hop-by-hop (routed)
PATHs and Data ====> Short-cut
Follow Short-cut
Path
Figure 2: Asymmetric RSVP Message Forwarding With ATM Short-CutsHere's the plot twist 🫢 RSVP already has mechanisms that support asymmetric paths. It's the same one used when RESV messages arrive at the wrong router and the wrong interface. RSVP messages are only processed when they arrive on the proper interface. When they arrive on the wrong interface, RSVP forwards them along. The proper interface is indicated in the NHOP object of the message. So the existing RSVP mechanisms already handle the asymmetric paths that can show up with short-cuts 🙌
Short-cut VC setup still raises a few issues when you run it with RSVP. The big ones are dealing with best effort short-cuts that are already established, deciding when to set up short-cuts, and QoS-only short-cuts. RSVP implementations will need to deal with these 🧩
The key issue for RSVP over ATM implementations is deciding when to set up a short-cut for a QoS data flow 💖 RSVP over ATM implementations SHOULD simply follow best effort traffic. When a short-cut has been set up for best effort traffic to a destination or next-hop, that same endpoint SHOULD be used when setting up RSVP triggered VCs for QoS traffic to the same destination or next-hop 🎀 This happens naturally when PATH messages are forwarded over the best effort short-cut ✨ Note that with this approach, if best effort short-cuts are never established, RSVP triggered QoS short-cuts will never get established either 🥲
2.4 Data VC Management for Heterogeneous Sessions 🌈
Heterogeneous sessions can only happen with multicast RSVP sessions 👑 The issues around data VC management for heterogeneous sessions are covered in detail in [6] and aren't repeated here. Quick recap: heterogeneity happens when receivers request different levels of QoS within a single session, and also when some receivers don't request any QoS at all 💅 Both types are shown in figure 3.
+----+
+------> | R1 |
| +----+
|
| +----+
+-----+ -----+ +--> | R2 |
| | ---------+ +----+ Receiver Request Types:
| Src | ----> QoS 1 and QoS 2
| | .........+ +----+ ....> Best-Effort
+-----+ .....+ +..> | R3 |
: +----+
/\ :
|| : +----+
|| +......> | R4 |
|| +----+
Single
IP Mulicast
Group
Figure 3: Types of Multicast Receivers[6] gives four models for dealing with heterogeneity: full heterogeneity, limited heterogeneity, homogeneous, and modified homogeneous models 🧠 The key issue for any implementation is providing the requested QoS downstream. One of these models, or some combination of them, from [6] may be used to provide the requested QoS. Unfortunately, none of the described models is the right answer for every case 😩 For some networks, e.g. public WANs, the limited heterogeneous model or a hybrid limited-full heterogeneous model will probably be wanted. In other networks, e.g. LANs, the modified homogeneous model will probably be wanted 🏠
Since there isn't one model that fits every case, implementations SHOULD implement either the limited heterogeneity model or the modified homogeneous model 💖 Implementations SHOULD support both approaches and give users the ability to pick which method is actually used, but they are not required to do so 🫶
3. Security Considerations 🚨
The same considerations stated in [4] and [8] apply to this document. There are no additional security issues raised in this document. 🔐
4. Acknowledgments 🫶
This work is based on earlier drafts and comments from the ISSLL working group. The author wants to thank them for their contribution, most notably Steve Berson, who coauthored one of the drafts 🌸
5. Author's Address 📍
Lou Berger
FORE Systems
1595 Spring Hill Road
5th Floor
Vienna, VA 22182
Phone: +1 703-245-4527
EMail: lberger@fore.comReferences 📚
okay so this is the reading list, bestie 🫶 every source the RFC leans on is here, with all the details so you can go find them yourself:
- [1] The ATM Forum, "MPOA Baseline Version 1", May 1997.
- [2] Berger, L., "RSVP over ATM Implementation Requirements", RFC 2380, August 1998.
- [3] Borden, M., and M. Garrett, "Interoperation of Controlled-Load and Guaranteed-Service with ATM", RFC 2381, August 1998.
- [4] Braden, R., Zhang, L., Berson, S., Herzog, S., and S. Jamin, "Resource ReSerVation Protocol (RSVP) -- Version 1 Functional Specification", RFC 2205, September 1997.
- [5] Bradner, S., "Key words for use in RFCs to Indicate Requirement Levels", BCP 14, RFC 2119, March 1997.
- [6] Crawley, E., Berger, L., Berson, S., Baker, F., Borden, M., and J. Krawczyk, "A Framework for Integrated Services and RSVP over ATM", RFC 2382, August 1998.
- [7] Luciani, J., Katz, D., Piscitello, D., and B. Cole, "NBMA Next Hop Resolution Protocol (NHRP)", RFC 2332, April 1998.
- [8] Perez, M., Liaw, F., Grossman, D., Mankin, A., Hoffman, E., and A. Malis, "ATM Signalling Support for IP over ATM", RFC 1755, February 1995.
Full Copyright Statement 🎀
the legal bit, and yes it matters, so read it all 🙈
Copyright (C) The Internet Society (1998). All Rights Reserved.
This document and translations of it may be copied and furnished to others, and derivative works that comment on or otherwise explain it or assist in its implementation may be prepared, copied, published and distributed, in whole or in part, without restriction of any kind, provided that the above copyright notice and this paragraph are included on all such copies and derivative works. However, this document itself may not be modified in any way, such as by removing the copyright notice or references to the Internet Society or other Internet organizations, except as needed for the purpose of developing Internet standards in which case the procedures for copyrights defined in the Internet Standards process must be followed, or as required to translate it into languages other than English.
The limited permissions granted above are perpetual and will not be revoked by the Internet Society or its successors or assigns.
This document and the information contained herein is provided on an "AS IS" basis and THE INTERNET SOCIETY AND THE INTERNET ENGINEERING TASK FORCE DISCLAIMS ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. 💅
📦 where this page comes from
- ietf: rfc2379 1998-08
- 💖 girlified by claude-haiku-5-5 on : a language model's rewrite of the original, under its terms