< draft-fang-mpls-hsdn-for-hsdc-04.txt   draft-fang-mpls-hsdn-for-hsdc-05.txt >
INTERNET-DRAFT Luyuan Fang INTERNET-DRAFT Luyuan Fang
Intended Status: Informational Deepak Bansal Intended Status: Informational Deepak Bansal
Expires: January 23, 2016 Microsoft Expires: July 26, 2016 Microsoft
Fabio Chiussi Fabio Chiussi
Chandra Ramachandran Chandra Ramachandran
Juniper Networks Juniper Networks
Ebben Aries Ebben Aries
Facebook Facebook
Shahram Davari Shahram Davari
Broadcom Broadcom
Barak Gafni Barak Gafni
Mellanox Mellanox
Daniel Voyer Daniel Voyer
Bell Canada Bell Canada
Nabil Bitar Nabil Bitar
Verizon Verizon
July 22, 2015 January 23, 2016
MPLS-Based Hierarchical SDN for Hyper-Scale DC/Cloud MPLS-Based Hierarchical SDN for Hyper-Scale DC/Cloud
draft-fang-mpls-hsdn-for-hsdc-04 draft-fang-mpls-hsdn-for-hsdc-05
Abstract Abstract
This document describes Hierarchical SDN (HSDN), an architectural This document describes Hierarchical SDN (HSDN), an architectural
solution to scale the Data Center (DC) and Data Center Interconnect solution to scale the Data Center (DC) and Data Center Interconnect
(DCI) networks to support tens of millions of physical underlay (DCI) networks to support tens of millions of physical underlay
endpoints, while efficiently handling both Equal Cost Multi Path endpoints, while efficiently handling both Equal Cost Multi Path
(ECMP) load-balanced traffic and any-to-any end-to-end Traffic (ECMP) load-balanced traffic and any-to-any end-to-end Traffic
Engineered (TE) traffic. HSDN achieves massive scale using Engineered (TE) traffic. HSDN achieves massive scale using
surprisingly small forwarding tables in the network nodes. HSDN surprisingly small forwarding tables in the network nodes. HSDN
skipping to change at page 2, line 21 skipping to change at page 2, line 21
material or to cite them other than as "work in progress." material or to cite them other than as "work in progress."
The list of current Internet-Drafts can be accessed at The list of current Internet-Drafts can be accessed at
http://www.ietf.org/1id-abstracts.html http://www.ietf.org/1id-abstracts.html
The list of Internet-Draft Shadow Directories can be accessed at The list of Internet-Draft Shadow Directories can be accessed at
http://www.ietf.org/shadow.html http://www.ietf.org/shadow.html
Copyright Notice Copyright Notice
Copyright (c) 2015 IETF Trust and the persons identified as the Copyright (c) 2016 IETF Trust and the persons identified as the
document authors. All rights reserved. document authors. All rights reserved.
This document is subject to BCP 78 and the IETF Trust's Legal This document is subject to BCP 78 and the IETF Trust's Legal
Provisions Relating to IETF Documents Provisions Relating to IETF Documents
(http://trustee.ietf.org/license-info) in effect on the date of (http://trustee.ietf.org/license-info) in effect on the date of
publication of this document. Please review these documents publication of this document. Please review these documents
carefully, as they describe your rights and restrictions with respect carefully, as they describe your rights and restrictions with respect
to this document. Code Components extracted from this document must to this document. Code Components extracted from this document must
include Simplified BSD License text as described in Section 4.e of include Simplified BSD License text as described in Section 4.e of
the Trust Legal Provisions and are provided without warranty as the Trust Legal Provisions and are provided without warranty as
skipping to change at page 5, line 42 skipping to change at page 5, line 42
nodes (e.g., at the Top-of-Rack (ToR) switches), or intermediate nodes (e.g., at the Top-of-Rack (ToR) switches), or intermediate
lookups in the network could be introduced, or even a combination of lookups in the network could be introduced, or even a combination of
MPLS and IP forwarding could be deployed as part of the HSDN network. MPLS and IP forwarding could be deployed as part of the HSDN network.
The HSDN underlay network is suited to support any Layer 2 or Layer 3 The HSDN underlay network is suited to support any Layer 2 or Layer 3
virtualized overlay network technology. In this document, we assume a virtualized overlay network technology. In this document, we assume a
MPLS-based overlay technology using a Virtual Network (VN) Label, MPLS-based overlay technology using a Virtual Network (VN) Label,
which is encapsulated in the HSDN label stack. However the which is encapsulated in the HSDN label stack. However the
description can be easily generalized to any overlay technology, such description can be easily generalized to any overlay technology, such
as BGP/MPLS IP VPNs [RFC4364], EVPN [RFC7432], VXLAN [RFC7348], NVGRE as BGP/MPLS IP VPNs [RFC4364], EVPN [RFC7432], VXLAN [RFC7348], NVGRE
[I-D.sridharan-virtualization-nvgre], Geneve [RFC7637], Geneve [I-D.draft-gross-geneve], and other technologies.
[I-D.draft-gross-geneve], and other technologies.
HSDN achieves massive scale using surprisingly small LFIBs in the HSDN achieves massive scale using surprisingly small LFIBs in the
network nodes, while supporting both ECMP load-balanced traffic and network nodes, while supporting both ECMP load-balanced traffic and
any-to-any end-to-end TE traffic [HSDNSOSR15]. HSDN also brings any-to-any end-to-end TE traffic [HSDNSOSR15]. HSDN also brings
important simplifications in the control plane and in the important simplifications in the control plane and in the
architecture of the SDN controller. architecture of the SDN controller.
The HSDN architecture and operation is characterized by two The HSDN architecture and operation is characterized by two
fundamental properties. First, all paths in the network are pre- fundamental properties. First, all paths in the network are pre-
established in the forwarding tables. Second, the HSDN labels can established in the forwarding tables. Second, the HSDN labels can
skipping to change at page 14, line 37 skipping to change at page 14, line 37
the one used for the partitions may be used. the one used for the partitions may be used.
o In Figure 3, we also show an additional Overlay Level. This o In Figure 3, we also show an additional Overlay Level. This
corresponds to the virtualized overlay network (if any) providing corresponds to the virtualized overlay network (if any) providing
Virtual Networks (VN) connecting Virtual Machines (VMs) and other Virtual Networks (VN) connecting Virtual Machines (VMs) and other
overlay network endpoints. Overlay network traffic is encapsulated overlay network endpoints. Overlay network traffic is encapsulated
by the HSDN underlay network. As mentioned in the Introduction, by the HSDN underlay network. As mentioned in the Introduction,
the HSDN underlay network is suited to support any Layer 2 or the HSDN underlay network is suited to support any Layer 2 or
Layer 3 virtualized overlay network technology, such as BGP/MPLS Layer 3 virtualized overlay network technology, such as BGP/MPLS
IP VPNs [RFC4364], EVPN [RFC7432], VXLAN [RFC7348], NVGRE IP VPNs [RFC4364], EVPN [RFC7432], VXLAN [RFC7348], NVGRE
[I-D.sridharan-virtualization-nvgre], Geneve [RFC7637], Geneve [I-D.draft-gross-geneve], and other
[I-D.draft-gross-geneve], and other technologies. A full technologies. A full description of the encapsulation of these
description of the encapsulation of these technologies into the technologies into the HSDN underlay label stack is out of scope of
HSDN underlay label stack is out of scope of this document and this document and will be addressed in a separate document.
will be addressed in a separate document.
The UPs are designed to contain one or more tiers of switches in the The UPs are designed to contain one or more tiers of switches in the
DC topology or nodes in the DCI. The key design criteria in defining DC topology or nodes in the DCI. The key design criteria in defining
the partitions at each level is that they need to follow the the partitions at each level is that they need to follow the
"natural" connectivity implemented in the DC/DCI topology. An example "natural" connectivity implemented in the DC/DCI topology. An example
is given in Section 3.2.3 to further clarify how the partitions are is given in Section 3.2.3 to further clarify how the partitions are
designed. designed.
3.2. Underlay Partition Border Nodes 3.2. Underlay Partition Border Nodes
Once the HSDN hierarchical partitioning is defined, Underlay Once the HSDN hierarchical partitioning is defined, Underlay
Partition Border Nodes (UPBNs) are assigned to each UP. This is Partition Border Nodes (UPBNs) are assigned to each UP. This is
illustrated in Figure 4. illustrated in Figure 4.
+++++++++++++++++++++++++++++++++++++++++++++++++ +++++++++++++++++++++++++++++++++++++++++++++++++
| UP0 <--------|-----+ | UP0 <--------|-----+
| | | | | |
| +++++++++++++++++++++++++++++++++++++++++++++ | | | +++++++++++++++++++++++++++++++++++++++++++++ | |
| | +-----------------------------------+ | | | | | +-----------------------------------+ | | |
| | | +---------+ UPBG1-i +---------+ | | | | | | | +---------+ UPBG1-i +---------+ | | | |
skipping to change at page 31, line 40 skipping to change at page 31, line 40
11.2 Informative References 11.2 Informative References
[RFC5920] Fang, L., Ed., "Security Framework for MPLS and GMPLS [RFC5920] Fang, L., Ed., "Security Framework for MPLS and GMPLS
Networks", RFC 5920, July 2010. Networks", RFC 5920, July 2010.
[RFC7348] M. Mahalingam et al., "Virtual eXtensible Local Area [RFC7348] M. Mahalingam et al., "Virtual eXtensible Local Area
Network (VXLAN): A Framework for Overlaying Virtualized Network (VXLAN): A Framework for Overlaying Virtualized
Layer 2 Networks over Layer 3 Networks", RFC 7348, August Layer 2 Networks over Layer 3 Networks", RFC 7348, August
2014. 2014.
[RFC7637] P. Garg et al., "NVGRE: Network Virtualization using
Generic Routing Encapsulation", RFC 7637, Sept. 2015.
[I-D.fang-idr-bgplu-for-hsdn] L. Fang et al., "BGP-LU for HSDN Label [I-D.fang-idr-bgplu-for-hsdn] L. Fang et al., "BGP-LU for HSDN Label
Distribution", draft-fang-idr-bgplu-for-hsdn-01 (work in Distribution", draft-fang-idr-bgplu-for-hsdn-02 (work in
progress), July 2015. progress), July 2015.
[I-D.sridharan-virtualization-nvgre] M. Sridharan et al., "NVGRE:
Network Virtualization using Generic Routing
Encapsulation", draft-sridharan-virtualization-nvgre-08
(work in progress), April 2015.
[I-D.draft-gross-geneve] J. Gross et al., "Geneve: Generic Network [I-D.draft-gross-geneve] J. Gross et al., "Geneve: Generic Network
Virtualization Encapsulation", draft-gross-geneve-02 (work Virtualization Encapsulation", draft-gross-geneve-02 (work
in progress), October 2014. in progress), October 2014.
[HSDNSOSR15] L. Fang et al., "Hierarchical SDN for the Hyper-Scale, [HSDNSOSR15] L. Fang et al., "Hierarchical SDN for the Hyper-Scale,
Hyper-Elastic Data Center and Cloud", ACM SIGCOMM Hyper-Elastic Data Center and Cloud", ACM SIGCOMM
Symposium on SDN Research 2015, Santa Clara, CA, June Symposium on SDN Research 2015, Santa Clara, CA, June
2015. 2015.
Authors' Addresses Authors' Addresses
 End of changes. 10 change blocks. 
17 lines changed or deleted 14 lines changed or added

This html diff was produced by rfcdiff 1.48. The latest version is available from http://tools.ietf.org/tools/rfcdiff/