Sep 16, 2007 10:18 PM
Mendel's continuous availability demo at VMworld
Mendel Rosenblum, VMware co-founder & chief scientist, gave a demo at VMworld of continuous availability using virtualization. Using the record/replay feature (seen in Workstation but not yet ESX Server), Mendel showed a 'slave' ESX Server synched to a first, with the secondary machine replicating the primary, down to every machine instruction and interruption.
A few views during the event:
VMTN Blog :
9:05 Continuous Availability. Machine dies, HA automatically restarts it before a human has to notice.
Last year we demo's Record/Replay -- record every instruction and interrupt, and then replay later.
Now as a tech preview in Workstation.
Now Continuous Availability. Have two machines, synchronizing continuously via record/replay -- when the primary goes down, the secondary machine keeps running and doesn't skip an instruction.
Third Demo: An experimental ESX running an Exchange VM being exercised by a test program. Turn on Continuous Availability -- now a second VM is mirroring every single action and state on the primary VM. When we pull the plug, the secondary machine will notice in about a second the secondary will take over with zero interruption. (cue big applause)
Scott Lowe :
Mendel next moved on to the idea of high availability in the datacenter. The idea of VMware HA allows us to recover VMs when a physical host fails. In his keynote, he alludes to the idea of detecting failures within the guest (the host was still running) and restarting it on another host. That's nice, but what is really powerful is what Mendel discusses next, and the idea of using record/replay as a high availability solution, known as "continuous availability." The idea here is that an execution stream for a VM is being captured and redirected to a secondary VM, live and in real-time. This creates two VMs in lockstep with each other. In the event of a hardware failure, the secondary VM will instantaneously fails over. (I suppose I don't need to tell anyone that this could make a good replication technology as well.) For the final demo this morning, they demonstrated the continuous availability functionality.
The demo involved a pair of virtualized Exchange servers with some clients running LoadSim to generate a load against the virtual Exchange servers. The setup created "mirrored VMs". As a demonstration of the technology, Mendel pulled the power plug on one of the ESX servers, and the secondary VM and clients barely even noticed the failure. Very nice technology!
What do you think? Science fiction? (VMware, as always, does not talk about future features or dates) Just another way of keeping a hot box on standby? Or the essential data center tool of the future -- keeping that hot standby in sync for any app at the CPU instruction level.
