One of the things that has always been missing from VMware’s High Availability solution stack is application awareness. As I explained in one of my earlier posts this is something that VMware is actively working on. Instead of creating a full App clustering level VMware decided to extend “VM Monitoring” and created an API to enable App level resiliency.
At VMworld I briefly sat down with Tom Stephens who is part of the Technical Marketing Team as an expert on HA and of course the recently introduced App Monitoring. Tom explained me what App Monitoring enables our partners to do and he used Symantec as the example. Symantec monitors the Application and all its associated services and ensure appropriate action is taken depending on the type of failure. Now keep in mind, it is still a single node so in case of OS maintenance their will be a short downtime. However, I personally feel that this does bridge a gap, this could add that extra 9 and that extra level of assurance your customer needs for his tier-1 app.
Not only will it react to a failover, but it also ensures for instance that all service are stopped and started in the correct order if and when needed. Now think about that for a second, you are doing maintenance during the weekend and need to reboot some of the Application Servers which are owned by someone else. This feature would enable you to reboot the machine and guarantee that the App will be started correctly as it knows the dependencies!
Tom recently published a great article about this new HA functionality and the key benefits of it, make sure you read it on the VMware Uptime blog!
I just figured it would use some functions of Hyperic, which VMware bought. I guess that’s not the case. In that case does this align with the vision of monitoring applications contained in vFabric?
Great post, I’ve bookmarked this page and have a feeling I’ll be returning to it frequently.