draft-ietf-nvo3-mcast-framework-03.txt   draft-ietf-nvo3-mcast-framework-04.txt 
skipping to change at page 1, line 15 skipping to change at page 1, line 15
Expires: August 14, 2016 M. McBride Expires: August 14, 2016 M. McBride
Huawei Huawei
V. Bannai V. Bannai
Google Google
R. Krishnan R. Krishnan
Dell Dell
February 15, 2016 February 15, 2016
A Framework for Multicast in NVO3 A Framework for Multicast in NVO3
draft-ietf-nvo3-mcast-framework-03 draft-ietf-nvo3-mcast-framework-04
Status of this Memo Status of this Memo
This Internet-Draft is submitted in full conformance with the This Internet-Draft is submitted in full conformance with the
provisions of BCP 78 and BCP 79. provisions of BCP 78 and BCP 79.
This Internet-Draft is submitted in full conformance with the This Internet-Draft is submitted in full conformance with the
provisions of BCP 78 and BCP 79. This document may not be modified, provisions of BCP 78 and BCP 79. This document may not be modified,
and derivative works of it may not be created, except to publish it and derivative works of it may not be created, except to publish it
as an RFC and to translate it into languages other than English. as an RFC and to translate it into languages other than English.
skipping to change at page 3, line 33 skipping to change at page 3, line 33
The reader is assumed to be familiar with the terminology as defined The reader is assumed to be familiar with the terminology as defined
in the NVO3 Framework document [RFC7365] and NVO3 Architecture in the NVO3 Framework document [RFC7365] and NVO3 Architecture
document [NVO3-ARCH]. document [NVO3-ARCH].
1.1. Infrastructure multicast 1.1. Infrastructure multicast
Infrastructure multicast includes protocols such as ARP/ND, DHCP, Infrastructure multicast includes protocols such as ARP/ND, DHCP,
and mDNS. It is possible to provide solutions for these that do not and mDNS. It is possible to provide solutions for these that do not
involve multicast in the underlay network. In the case of ARP/ND, involve multicast in the underlay network. In the case of ARP/ND,
an NVA can be used for distributing the mappings of IP address to an NVA can be used for distributing the mappings of IP address to
MAC address to all NVEs. The NVEs can then trap ARP Request/ND MAC address to all NVEs. The NVEs can then trap ARP Request/ND
Neighbor Solicitation messages from the TSs that are attached to it Neighbor Solicitation messages from the TSs that are attached to it
and respond to them, thereby eliminating the need to for and respond to them, thereby eliminating the need to for
broadcast/multicast of such messages. In the case of DHCP, the NVE broadcast/multicast of such messages. In the case of DHCP, the NVE
can be configured to forward these messages using a helper function. can be configured to forward these messages using a helper function.
Of course it is possible to support all of these infrastructure Of course it is possible to support all of these infrastructure
multicast protocols natively if the underlay provides multicast multicast protocols natively if the underlay provides multicast
transport. However, even in the presence of multicast transport, it transport. However, even in the presence of multicast transport, it
may be beneficial to use the optimizations mentioned above to reduce may be beneficial to use the optimizations mentioned above to reduce
the amount of such traffic in the network. the amount of such traffic in the network.
skipping to change at page 4, line 43 skipping to change at page 4, line 43
NVGRE: Network Virtualization using GRE NVGRE: Network Virtualization using GRE
SSM: Source-Specific Multicast SSM: Source-Specific Multicast
STT: Stateless Tunnel Transport STT: Stateless Tunnel Transport
TS: Tenant system TS: Tenant system
VM: Virtual Machine VM: Virtual Machine
VN: Virtual Network
VXLAN: Virtual eXtensible LAN VXLAN: Virtual eXtensible LAN
3. Multicast mechanisms in networks that use NVO3 3. Multicast mechanisms in networks that use NVO3
In NVO3 environments, traffic between NVEs is transported using an In NVO3 environments, traffic between NVEs is transported using an
encapsulation such as VXLAN [VXLAN], NVGRE [RFC7637], STT [STT], encapsulation such as VXLAN [VXLAN], NVGRE [RFC7637], STT [STT],
etc. etc.
Besides the need to support the Address Resolution Protocol (ARP) Besides the need to support the Address Resolution Protocol (ARP)
and Neighbor Discovery (ND), there are several applications that and Neighbor Discovery (ND), there are several applications that
skipping to change at page 6, line 32 skipping to change at page 6, line 32
service without requiring any specific support from the underlay, service without requiring any specific support from the underlay,
other than that of a unicast service. A multicast or broadcast other than that of a unicast service. A multicast or broadcast
transmission is achieved by replicating the packet at the source transmission is achieved by replicating the packet at the source
NVE, and making copies, one for each destination NVE that the NVE, and making copies, one for each destination NVE that the
multicast packet must be sent to. multicast packet must be sent to.
For this mechanism to work, the source NVE must know, a priori, the For this mechanism to work, the source NVE must know, a priori, the
IP addresses of all destination NVEs that need to receive the IP addresses of all destination NVEs that need to receive the
packet. For the purpose of ARP/ND, this would involve knowing the packet. For the purpose of ARP/ND, this would involve knowing the
IP addresses of all the NVEs that have Tenant Systems in the virtual IP addresses of all the NVEs that have Tenant Systems in the virtual
network instance (VNI) of the Tenant System that generated the network (VN) of the Tenant System that generated the request.
request. For the support of application-specific multicast traffic, For the support of application-specific multicast traffic,
a method similar to that of receiver-sites registration for a a method similar to that of receiver-sites registration for a
particular multicast group described in [LISP-Signal-Free] can be particular multicast group described in [LISP-Signal-Free] can be
used. The registrations from different receiver-sites can be merged used. The registrations from different receiver-sites can be merged
at the NVA, which can construct a multicast replication-list at the NVA, which can construct a multicast replication-list
inclusive of all NVEs to which receivers for a particular multicast inclusive of all NVEs to which receivers for a particular multicast
group are attached. The replication-list for each specific multicast group are attached. The replication-list for each specific multicast
group is maintained by the NVA. group is maintained by the NVA.
The receiver-sites registration is achieved by egress NVEs The receiver-sites registration is achieved by egress NVEs
performing the IGMP/MLD snooping to maintain state for which performing the IGMP/MLD snooping to maintain state for which
skipping to change at page 7, line 10 skipping to change at page 7, line 10
group. When the members of a multicast group are outside the NVO3 group. When the members of a multicast group are outside the NVO3
domain, it is necessary for NVO3 gateways to keep track of the domain, it is necessary for NVO3 gateways to keep track of the
remote members of each multicast group. The NVEs and NVO3 gateways remote members of each multicast group. The NVEs and NVO3 gateways
then communicate the multicast groups that are of interest to the then communicate the multicast groups that are of interest to the
NVA. If the membership is not communicated to the NVA, and if it is NVA. If the membership is not communicated to the NVA, and if it is
necessary to prevent hosts attached to an NVE that have not necessary to prevent hosts attached to an NVE that have not
subscribed to a multicast group from receiving the multicast subscribed to a multicast group from receiving the multicast
traffic, the NVE would need to maintain multicast group membership traffic, the NVE would need to maintain multicast group membership
information. information.
In multi-homing environments, i.e. in those where a TS is attached In the absence of IGMP/MLD snooping, the traffic would be delivered
to all hosts that are part of the VN.
In multi-homing environments, i.e., in those where a TS is attached
to more than one NVE, the NVA would be expected to provide to more than one NVE, the NVA would be expected to provide
information to all of the NVEs under its control about all of the information to all of the NVEs under its control about all of the
NVEs to which such a TS is attached. The ingress NVE can choose any NVEs to which such a TS is attached. The ingress NVE can choose any
one of the egress NVEs for the data frames destined towards the TS. one of the egress NVEs for the data frames destined towards the TS.
In the absence of IGMP/MLD snooping, the traffic would be delivered
to all hosts that are part of the VNI.
This method requires multiple copies of the same packet to all NVEs This method requires multiple copies of the same packet to all NVEs
that participate in the VN. If, for example, a tenant subnet is that participate in the VN. If, for example, a tenant subnet is
spread across 50 NVEs, the packet would have to be replicated 50 spread across 50 NVEs, the packet would have to be replicated 50
times at the source NVE. This also creates an issue with the times at the source NVE. This also creates an issue with the
forwarding performance of the NVE. forwarding performance of the NVE.
Note that this method is similar to what was used in VPLS [RFC4792] Note that this method is similar to what was used in VPLS [RFC4762]
prior to support of MPLS multicast [RFC7117]. While there are some prior to support of MPLS multicast [RFC7117]. While there are some
similarities between MPLS VPN and the NVO3 overlay, there are some similarities between MPLS VPN and the NVO3 overlay, there are some
key differences: key differences:
- The CE-to-PE attachment in VPNs is somewhat static, whereas in a - The CE-to-PE attachment in VPNs is somewhat static, whereas in a
DC that allows VMs to migrate anywhere, the TS attachment to NVE DC that allows VMs to migrate anywhere, the TS attachment to NVE
is much more dynamic. is much more dynamic.
- The number of PEs to which a single VPN customer is attached in - The number of PEs to which a single VPN customer is attached in
an MPLS VPN environment is normally far less than the number of an MPLS VPN environment is normally far less than the number of
NVEs to which a VNI's VMs are attached in a DC. NVEs to which a VN's VMs are attached in a DC.
When a VPN customer has multiple multicast groups, [RFC6513] When a VPN customer has multiple multicast groups, [RFC6513]
"Multicast VPN" combines all those multicast groups within each "Multicast VPN" combines all those multicast groups within each
VPN client to one single multicast group in the MPLS (or VPN) VPN client to one single multicast group in the MPLS (or VPN)
core. The result is that messages from any of the multicast core. The result is that messages from any of the multicast
groups belonging to one VPN customer will reach all the PE nodes groups belonging to one VPN customer will reach all the PE nodes
of the client. In other words, any messages belonging to any of the client. In other words, any messages belonging to any
multicast groups under customer X will reach all PEs of the multicast groups under customer X will reach all PEs of the
customer X. When the customer X is attached to only a handful of customer X. When the customer X is attached to only a handful of
PEs, the use of this approach does not result in excessive wastage PEs, the use of this approach does not result in excessive wastage
of bandwidth in the provider's network. of bandwidth in the provider's network.
In a DC environment, a typical server/hypervisor based virtual In a DC environment, a typical server/hypervisor based virtual
switch may only support 10's VMs (as of this writing). A subnet switch may only support 10's VMs (as of this writing). A subnet
with N VMs may be, in the worst case, spread across N vSwitches. with N VMs may be, in the worst case, spread across N vSwitches.
Using "MPLS VPN multicast" approach in such a scenario would Using "MPLS VPN multicast" approach in such a scenario would
require the creation of a Multicast group in the core for this VNI require the creation of a Multicast group in the core for this VN
to reach all N NVEs. If only small percentage of this client's VMs to reach all N NVEs. If only small percentage of this client's VMs
participate in application specific multicast, a great number of participate in application specific multicast, a great number of
NVEs will receive multicast traffic that is not forwarded to any NVEs will receive multicast traffic that is not forwarded to any
of their attached VMs, resulting in considerable wastage of of their attached VMs, resulting in considerable wastage of
bandwidth. bandwidth.
Therefore, the Multicast VPN solution may not scale in DC Therefore, the Multicast VPN solution may not scale in DC
environment with dynamic attachment of Virtual Networks to NVEs and environment with dynamic attachment of Virtual Networks to NVEs and
greater number of NVEs for each virtual network. greater number of NVEs for each virtual network.
skipping to change at page 9, line 14 skipping to change at page 9, line 14
- The MSN can obtain the membership information from the NVEs that - The MSN can obtain the membership information from the NVEs that
snoop the IGMP/MLD messages. This can be done by having the MSN snoop the IGMP/MLD messages. This can be done by having the MSN
communicate with the NVEs, or by having the NVA obtain the communicate with the NVEs, or by having the NVA obtain the
information from the NVEs, and in turn have MSN communicate with information from the NVEs, and in turn have MSN communicate with
the NVA. the NVA.
Unlike the method described in Section 3.2, there is no performance Unlike the method described in Section 3.2, there is no performance
impact at the ingress NVE, nor are there any issues with multiple impact at the ingress NVE, nor are there any issues with multiple
copies of the same packet from the source NVE to the multicast copies of the same packet from the source NVE to the multicast
service node. However there remain issues with multiple copies of service node. However, there remain issues with multiple copies of
the same packet on links that are common to the paths from the MSN the same packet on links that are common to the paths from the MSN
to each of the egress NVEs. Additional issues that are introduced to each of the egress NVEs. Additional issues that are introduced
with this method include the availability of the MSN, methods to with this method include the availability of the MSN, methods to
scale the services offered by the MSN, and the sub-optimality of the scale the services offered by the MSN, and the sub-optimality of the
delivery paths. delivery paths.
Finally, the IP address of the source NVE must be preserved in Finally, the IP address of the source NVE must be preserved in
packet copies created at the multicast service node if data plane packet copies created at the multicast service node if data plane
learning is in use. This could create problems if IP source address learning is in use. This could create problems if IP source address
reverse path forwarding (RPF) checks are in use. reverse path forwarding (RPF) checks are in use.
skipping to change at page 9, line 39 skipping to change at page 9, line 39
NVE encapsulates the packet with the appropriate IP multicast NVE encapsulates the packet with the appropriate IP multicast
address in the tunnel encapsulation header for delivery to the address in the tunnel encapsulation header for delivery to the
desired set of NVEs. The protocol in the underlay could be any desired set of NVEs. The protocol in the underlay could be any
variant of Protocol Independent Multicast (PIM), or protocol variant of Protocol Independent Multicast (PIM), or protocol
dependent multicast, such as [ISIS-Multicast]. dependent multicast, such as [ISIS-Multicast].
If an NVE connects to its attached TSs via Layer 2 network, there If an NVE connects to its attached TSs via Layer 2 network, there
are multiple ways for NVEs to support the application specific are multiple ways for NVEs to support the application specific
multicast: multicast:
- The NVE only supports the basic IGMP/MLD snooping function, let - The NVE only supports the basic IGMP/MLD snooping function, let
the TSs routers handling the application specific multicast. This the TSs routers handling the application specific multicast. This
scheme doesn't utilize the underlay IP multicast protocols. scheme doesn't utilize the underlay IP multicast protocols.
- The NVE can act as a pseudo multicast router for the directly - The NVE can act as a pseudo multicast router for the directly
attached VMs and support proper mapping of IGMP/MLD's messages to attached VMs and support proper mapping of IGMP/MLD's messages to
the messages needed by the underlay IP multicast protocols. the messages needed by the underlay IP multicast protocols.
With this method, there are none of the issues with the methods With this method, there are none of the issues with the methods
described in Sections 3.2. described in Sections 3.2.
With PIM Sparse Mode (PIM-SM), the number of flows required would be With PIM Sparse Mode (PIM-SM), the number of flows required would be
(n*g), where n is the number of source NVEs that source packets for (n*g), where n is the number of source NVEs that source packets for
the group, and g is the number of groups. Bidirectional PIM (BIDIR- the group, and g is the number of groups. Bidirectional PIM (BIDIR-
PIM) would offer better scalability with the number of flows PIM) would offer better scalability with the number of flows
skipping to change at page 13, line 10 skipping to change at page 13, line 10
application specific multicast in networks that use NVO3. It application specific multicast in networks that use NVO3. It
highlights the basics of each mechanism and some of the issues with highlights the basics of each mechanism and some of the issues with
them. As solutions are developed, the protocols would need to them. As solutions are developed, the protocols would need to
consider the use of these mechanisms and co-existence may be a consider the use of these mechanisms and co-existence may be a
consideration. It also highlights some of the requirements for consideration. It also highlights some of the requirements for
supporting multicast applications in an NVO3 network. supporting multicast applications in an NVO3 network.
7. Security Considerations 7. Security Considerations
This draft does not introduce any new security considerations beyond This draft does not introduce any new security considerations beyond
what may be present in proposed solutions what may be present in proposed solutions.
8. IANA Considerations 8. IANA Considerations
This document requires no IANA actions. RFC Editor: Please remove This document requires no IANA actions. RFC Editor: Please remove
this section before publication. this section before publication.
9. References 9. References
9.1. Normative References 9.1. Normative References
skipping to change at page 14, line 15 skipping to change at page 14, line 15
[STT] Davie, B. and Gross, J., "A stateless transport tunneling [STT] Davie, B. and Gross, J., "A stateless transport tunneling
protocol for network virtualization," work in progress. protocol for network virtualization," work in progress.
[DC-MC] McBride, M. and Lui, H., "Multicast in the data center [DC-MC] McBride, M. and Lui, H., "Multicast in the data center
overview," work in progress. overview," work in progress.
[ISIS-Multicast] [ISIS-Multicast]
Yong, L. et al., "ISIS Protocol Extension for Building Yong, L. et al., "ISIS Protocol Extension for Building
Distribution Trees", work in progress. Distribution Trees", work in progress.
[RFC4792] Lasserre, M., and Kompella, V. (Eds.), "Virtual Private [RFC4762] Lasserre, M., and Kompella, V. (Eds.), "Virtual Private
LAN Service (VPLS) using Label Distribution Protocol (LDP) LAN Service (VPLS) using Label Distribution Protocol (LDP)
signaling," RFC 4762, January 2007. signaling," January 2007.
[RFC7117] Aggarwal, R. et al., "Multicast in VPLS," February 2014. [RFC7117] Aggarwal, R. et al., "Multicast in VPLS," February 2014.
[LANE] "LAN emulation over ATM," The ATM Forum, af-lane-0021.000, [LANE] "LAN emulation over ATM," The ATM Forum, af-lane-0021.000,
January 1995. January 1995.
[EDGE-REP] [EDGE-REP]
Marques P. et al., "Edge multicast replication for BGP IP Marques P. et al., "Edge multicast replication for BGP IP
VPNs," work in progress.. VPNs," work in progress..
 End of changes. 16 change blocks. 
18 lines changed or deleted 20 lines changed or added

This html diff was produced by rfcdiff 1.42. The latest version is available from http://tools.ietf.org/tools/rfcdiff/