I expect to hear a lot about the “wonderful” idea of moving running VMs 100 msec away (across the continent) in the upcoming weeks. I would recommend you read a few of my older blog posts before considering it… and don’t waste time trying to persuade the true believers with technical arguments – talk with whoever will foot the bill or walk away.
- Hot and cold VM mobility
- vMotion: an elephant in the data center
- Live VM mobility network requirements
- Video: networking requirements for VM mobility
- vMotion and VXLAN
Technical details (= problems) of long-distance vMotion and stretched VLANs
- Long-distance vMotion and the traffic trombone
- Migrating a cold VM into a foreign subnet
- Layer-3 gurus asleep at the wheel
- Layer-2 extension use cases
- Revisited: layer-2 DCI over VXLAN
- External routing with layer-2 DCI
Why long-distance vMotion doesn’t make sense
- High-availability fallacies
- Long-distance vMotion, stretched HA clusters and business needs
- Long-distance vMotion for disaster avoidance? Do the math!
- Long-distance workload mobility in perspective
- Coping with long-distance vMotion requests
- Follow-the-sun workload mobility? Get lost!
- Busting layer-2 DCI myths
- Virtualized squashed complexity sausage