Il y a a peu de questions that J'ai found on ServerFault that hint around this topic, et tandis que it may be somewhat opinion-based, Je pense it can fall into that "good subjective" category based on le below:
*Constructive subjective questions:*
```
* tend to have long, not short, answers
* have a constructive, fair, and impartial tone
* invite sharing experiences over opinions
* insist that opinion be backed up with facts and references
* are more than just mindless social fun
```
So that out of le way.
Je suis helping out a fellow sysadmin that is replacing an older physical server running Windows 2003 et he's looking to pas seulement replace le hardware mais "upgrade" to 2012 R2 in le process.
In our discussions about his replacement hardware, we discussed le possibility of him installing ESXi et alors making le 2012 "server" a VM et migrating le old apps/files/roles depuis le 2003 server to le VM à la place of to a non-VM install on le nouveau hardware.
He ne perceive tout time in le suivant peu de years le need to move anything else to a VM ou créez undditional VMs, so in le end this will soit be nouveau hardware running a normal install ou nouveau hardware running a single VM on ESXi.
My own experience would lean towards a VM still, there n'est pas a truly compelling reason to do so autre than possibilities that may arise to créez undditional VMs. But there is le additional overhead et management aspect of le hypervisor now, albeit J'ai experienced better management capabilities et reporting capabilities avec a VM.
So avec le premise of hoping this can stay in le "good subjective" category to help others in le future, **what experiences/facts/references/constructive answers do you have to help support soit outcome (virtualizing ou pas a single "server")?**
Lors du déploiement d'un seul serveur sur du nouveau matériel, faut-il le virtualiser ?
Re: Lors du déploiement d'un seul serveur sur du nouveau matériel, faut-il le virtualiser ?
In le general case, le advantage of putting a standalone server on a hypervisor is future-proofing. It makes future expansion ou upgrades much easier, much faster, et en conséquence, cheaper. The primary drawback is additional complexity et cost (not necessarily financially, mais depuis a man-hours et time perspective).
So, to come to a decision, I ask myself three questions (and usually prefer to put le server on a hypervisor, for what c'est worth).
- How big is le added cost of le hypervisor?
- Financially, c'est *usually* minimal ou non-existent.
- Both VMware et Microsoft have licensing options that allow you to run a host et a single guest for free, et this is sufficient for le plus standalone servers, exceptions generally being servers that are especially resource-intensive.
- From a management et resource standpoint, determining cost can be a bit trickier.
- You basically double le cost of maintaining le system, parce que now you have two systems to monitor, manage et keep up-to-date avec patches et updates (the guest OS et le host OS).
- For le plus uses, this is pas a big deal, as c'est pas terribly taxing to maintain one server, bien que for certains especially small ou especially technically challenged organizations, this can be a real concern.
- You aussi add to le technical skills required. Now, à la place of juste needing someone who can download updates depuis Windows Update, you need someone who knows enough to manage et maintain le virtualization environment.
- Again, pas usually a problem, mais sometimes, c'est more than an organization can handle.
- How big is le benefit depuis ease-of upgrade ou expansion?
- This boils down to how likely future expansion is, parce que obviously, si they ne expand ou upgrade leur server assets, this benefit is zero.
- If this is le type of organization c'est juste going to stuff le server in a corner et forget about it for 10 years jusqu'à it needs to be replaced anyway, il y a no point.
- If they're likely to grow organizationally, ou even juste technically (by say adding nouveau servers avec différent roles, à la place of juste having an all-in-one server), alors this provides a fairly substantial benefit.
- Quel est le benefit *now*?
- Virtualization bring benefits beyond future-proofing, et in certains use-cases, they can be substantial.
- The le plus obvious one is le ability to create snapshots et trivial-to-restore backups avant doing something on le system, so si it goes bad, you can revert in one click.
- The ability to experiment avec autre VMs (and play le "what if" game) is another one J'ai seen management get excited about.
For mon money, though, le biggest benefit is le added portability you get depuis running a production server on a hypervisor. If something goes really wrong et you get yourself into a disaster-recovery ou restore-from-backups situation, it is almost infinitely easier to restore a disk image to a machine running le même hypervisor than trying to do a bare-metal restore.
So, to come to a decision, I ask myself three questions (and usually prefer to put le server on a hypervisor, for what c'est worth).
- How big is le added cost of le hypervisor?
- Financially, c'est *usually* minimal ou non-existent.
- Both VMware et Microsoft have licensing options that allow you to run a host et a single guest for free, et this is sufficient for le plus standalone servers, exceptions generally being servers that are especially resource-intensive.
- From a management et resource standpoint, determining cost can be a bit trickier.
- You basically double le cost of maintaining le system, parce que now you have two systems to monitor, manage et keep up-to-date avec patches et updates (the guest OS et le host OS).
- For le plus uses, this is pas a big deal, as c'est pas terribly taxing to maintain one server, bien que for certains especially small ou especially technically challenged organizations, this can be a real concern.
- You aussi add to le technical skills required. Now, à la place of juste needing someone who can download updates depuis Windows Update, you need someone who knows enough to manage et maintain le virtualization environment.
- Again, pas usually a problem, mais sometimes, c'est more than an organization can handle.
- How big is le benefit depuis ease-of upgrade ou expansion?
- This boils down to how likely future expansion is, parce que obviously, si they ne expand ou upgrade leur server assets, this benefit is zero.
- If this is le type of organization c'est juste going to stuff le server in a corner et forget about it for 10 years jusqu'à it needs to be replaced anyway, il y a no point.
- If they're likely to grow organizationally, ou even juste technically (by say adding nouveau servers avec différent roles, à la place of juste having an all-in-one server), alors this provides a fairly substantial benefit.
- Quel est le benefit *now*?
- Virtualization bring benefits beyond future-proofing, et in certains use-cases, they can be substantial.
- The le plus obvious one is le ability to create snapshots et trivial-to-restore backups avant doing something on le system, so si it goes bad, you can revert in one click.
- The ability to experiment avec autre VMs (and play le "what if" game) is another one J'ai seen management get excited about.
For mon money, though, le biggest benefit is le added portability you get depuis running a production server on a hypervisor. If something goes really wrong et you get yourself into a disaster-recovery ou restore-from-backups situation, it is almost infinitely easier to restore a disk image to a machine running le même hypervisor than trying to do a bare-metal restore.