Hi, Yes, Major drawback of GRE kind of enacp, switch need to parse deeply into GRE packets for applying policies related to load distribution Port Channels that are used to aggregate the bandwidth of multiple physical links into one logical link.
If switches were to try to distribute GRE flows between two VTEPs that used a GRE encapsulation, all the traffic would be directed to use only one link within these Port Channels. However Parsing all the way to UDP (ex VxLan) source and destination port numbers, (may be configure the switches to use 5-tuple) can spread each UDP flow out to a different link of a Port Channel or ECMP route. Regards, keshava -----Original Message----- From: [email protected] [mailto:[email protected]] On Behalf Of Xuxiaohu Sent: Tuesday, May 08, 2012 8:51 AM To: [email protected]; [email protected]; [email protected] Subject: [nvo3] Facilitating the load-balancing of L2VPN/L3VPN traffic over IP PSN using MPLS-in-UDP encapsulation// fwd: New Version Notification for draft-xu-mpls-in-udp-00.txt Hi all, This topic may be interesting to L2VPN, L3VPN and even NVo3 folks. Any comments are welcome. Best regards, Xiaohu/ Marshall/Lucy > -----邮件原件----- > 发件人: Xuxiaohu > 发送时间: 2012年4月28日 10:55 > 收件人: '[email protected]' > 抄送: '[email protected]'; Lucy yong > 主题: fwd: New Version Notification for draft-xu-mpls-in-udp-00.txt > > Hi all, > > Equal Cost Multi-Path (ECMP) and Link Aggregation Group (LAG) are widely used > in the core of IP-enabled Packet Switch Networks (PSN) for load-balancing > purposes. Most core routers (i.e., P routers) in the IP-enabled PSN are > capable of > load-balancing IP traffic flows across ECMP paths and/or LAG based on the hash > of the five-tuple of UDP/TCP packets (i.e., source IP address, destination > IP > address, source port, destination port, and protocol) or some fields in the IP > header of non-UDP/TCP packets (e.g., source IP address, destination IP > address). > > However, with existing IP-based encapsulation methods as defined in [RFC4023] > such as MPLS-in-IP and MPLS-in-GRE, distinct customer traffic flows of various > MPLS applications (e.g., MPLS-based L2VPN or L3VPN) between a given PE pair > would be encapsulated with the same IP or GRE tunnel header prior to > traversing > the core. Since the encapsulating traffic is neither TCP nor UDP traffic, core > routers could only perform hash calculation on the fields in the IP header of > IP or > GRE tunnels. As a result, core routers could not achieve an effective > load-balancing for these traffic flows in the network due to the lack of > adequate > entropy information. > > This document specifies one additional IP-based encapsulation technology for > MPLS packets referred to as MPLS-in-UDP, which is intended to facilitate > load-balancing the traffic of various MPLS applications such as MPLS-based > L2VPN > and L3VPN in the core of IP-enabled PSN. > > Any comments are welcome. > > Best regards, > Xiaohu > > -----邮件原件----- > 发件人: [email protected] [mailto:[email protected]] > 发送时间: 2012年4月28日 10:18 > 收件人: Xuxiaohu > 抄送: Lucy yong; [email protected] > 主题: New Version Notification for draft-xu-mpls-in-udp-00.txt > > A new version of I-D, draft-xu-mpls-in-udp-00.txt has been successfully > submitted by Xiaohu Xu and posted to the IETF repository. > > Filename: draft-xu-mpls-in-udp > Revision: 00 > Title: Encapsulating MPLS in UDP > Creation date: 2012-04-28 > WG ID: Individual Submission > Number of pages: 7 > > Abstract: > This document specifies one additional IP-based encapsulation > technology for MPLS packets referred to as MPLS-in-UDP, which is > intended to facilitate load-balancing the traffic of various MPLS > applications such as MPLS-based L2VPN and L3VPN in the core of IP- > enabled packet switch networks. > > > > > > The IETF Secretariat _______________________________________________ nvo3 mailing list [email protected] https://www.ietf.org/mailman/listinfo/nvo3 _______________________________________________ nvo3 mailing list [email protected] https://www.ietf.org/mailman/listinfo/nvo3
