Category: data center

How would you implement redundant VXLAN-based Data Center Interconnect (two DCs connected via 2 links terminated by 2 VTEPs in each location).

Could you do it without EVPN-type dual homing? Some vendors are proposing VXLAN+MLAG, but MLAG is a bad kludge prone to bugs (as per your comments). Would you use VXLAN + STP?

What device/technology would you select for a multi-tenant DC edge? In Cisco terms:

1. Use n9k (non aci) as circuit aggr + L3 border –> connect by mpls-sr to leaf spine fabric or just integrated? (Border leaf)
2. Use ASR routers (full netflow capabilities) as L3 border –> connect by mpls-sr to leaf spine fabric

Can we use VXLAN with EVPN to implement a data center interconnect (DCI) between two data centers? Can we use it to reduce the failover time to below 50 msec?

I’d be interested in seeing a updated/modernized DC internet edge design session, including some of the following topics:

• Carrier path selection and failover (AS-path based, performance/quality based, etc.)
• Path visibility (Tools like ThousandEyes, et al.)
• Public IP mobility and failover between DC’s (BGP strategy, iBGP topology and transport)
• Designing and building backbone connectivity between INET edges @ both DC’s (shared vs dedicated transport, routed vs. stretched segments, etc.)
• DDoS mitigation (BGP, DNS-based techniques, etc.)
• Impacts of IPv6 on all above topics (Design, hardware resources, platform selection, etc.)