parallels vs bootcamp performance

Discussion in 'General Questions' started by Richard_Lee1, Oct 10, 2014.

Thread Status:
Not open for further replies.
  1. Richard_Lee1

    Richard_Lee1 Junior Member

    Messages:
    18
    I heard it mentioned that a bootcamp partition running from within parallels is faster than a VM created within parallels.

    Can anybody comment on their experience?

    TIA
    Rich
     
  2. Abdul@Parallels

    Abdul@Parallels Parallels Support

    Messages:
    597
    Hi Rich,

    Parallels Desktop is a Virtualization software, which helps you to run Windows or other operating system in your Mac simultaneously.
    Many people aren’t aware of the differences between Boot Camp and Parallels Desktop. Although both allow you to run Windows on a Mac, that’s where the similarities end. Take your “Windows on a Mac” experience to the next level!
    Use Boot Camp in Parallels Desktop 10 for Mac and run Mac and Windows applications side-by-side, without rebooting.
    For more information, please refer to http://www.parallels.com/landingpage/pd/bootcamp/

    As far as there are no major difference between running OS in boot-camp or in Parallels Desktop unless your Mac OS has good configuration.
    Please try running boot-camp OS in Parallels Desktop : http://kb.parallels.com/112941 and share us your experience.
     
  3. Specimen

    Specimen Product Expert

    Messages:
    3,242
    Under Parallels being Bootcamp or a virtual drive shouldn't make a noticeable difference, and I would argue they have about the same R&W speeds, however you will notice a difference if the virtual disk or Bootcamp partition is in a different physical disk than OS X, for a very good reason, the VM will not compete for I/O with OS X.

    I hope that helps.
     
  4. Richard_Lee1

    Richard_Lee1 Junior Member

    Messages:
    18
    Not really.

    Minor point: The term "virtual drive" typically refers to a mounted cd/dvd/ or iso, and not to a "virtual machine"

    Major point: Where you say a difference may be noted where a VM is on a separate physical disk then I _think_ you're saying that the VM on a separate disk would be faster than bootcamp on the primary disk. Do you have anything to back that up? I think it's well-known that running a VM on a separate spindle is faster than running *the VM* on the same spindle as the host OS, but faster than bootcamp alone? It would be really interesting to know - maybe backed up with some in-house test results, and a view of how IO is routed under the different configurations.

    Thanks!
     
  5. Richard_Lee1

    Richard_Lee1 Junior Member

    Messages:
    18
    Wow.

    So my question was "does the bootcamp OS run faster than the VM in parallels?" Your answer is to tell me what Parallels does (uh, thanks), try to *sell* me Parallels (I already bought it - your efforts don't make me want to renew), tell me how to run bootcamp (using BootCamp *in* Parallels isn't the way I'd describe it btw)... then you tell me there are no differences but suggest I try it and let you know!

    You make one interesting point - based on "good configuration", but you entirely fail to explain what the configuration differences are that might enable me to get the best out of Parallels on my Mac.

    This is such a random response Abdul. I continue to be impressed with Parallels and entirely unimpressed with the marketing and support, which really do a great disservice to the product and the engineers who build it.
     
  6. Specimen

    Specimen Product Expert

    Messages:
    3,242
    Let me clarify for you what 'Virtual Drive'/'Virtual HardDrive'/'Virtual HardDisk'/'Virtual HDD' means. Virtual Drive refers to the .hds/hdd file residing inside the .pvm package of a virtual machine, it's 'virtual' as opposed to 'real/physical' because the whole 'hard disk/partition' is a file. In a similar way an .iso or .dmg file (image files) can be mounted as a virtual dvd.
    But I'm sure you are not interested in hair splitting.

    Now that's out of the way, let's move on to the major one...
    What you wrote in your initial post is about comparing I/O performance under Parallels of a VM based on Bootcamp versus a VM based on a Virtual Disk ("from within parallels"), and NOT comparing the performance running the Bootcamp partition natively (Booting the mac to Windows). Parallels can use a Bootcamp partition as the storage/disk of a VM.
    And in that scenario you really don't need benchmarks to realise that if both OS X and the Windows VM, either in a Bootcamp partition or in a Virtual Disk/Drive, reside on the same physical disk they will be competing for I/O the moment they'll both want to read and write something from the disk at the same time, hence the slower performance.
     
    Last edited: Oct 26, 2014
  7. Specimen

    Specimen Product Expert

    Messages:
    3,242
    "does the bootcamp OS run faster than the VM in parallels?"

    Mind you that this not what you asked initially, you specifically said "from within Parallels" (Parallels can use a Bootcamp partition as the storage/disk of a VM).

    The answer to the question if booting natively into bootcamp has a better performance than booting via Parallels, the answer is, obviously that the performance is better if you boot into Bootcamp natively as you don't have OS X using resources, simple as that, and Windows talks directly with the hardware and has the whole hardware for itself.

    Your attitude towards Abdul and me seems very aggressive and gratuitous, specially considering your question is unclear and apparently it's not even what you want to know.

    If you want an informative and informed reply start with an informed and informative question.
     
    Last edited: Oct 26, 2014
  8. Richard_Lee1

    Richard_Lee1 Junior Member

    Messages:
    18
    It's not hair-splitting, and I don't welcome your attitude. Defining "Virtual" when the point was about "Machine/Disk" is patronising. From a typical user's point of view we're talking about Virtual Machines - that's what Parallels exposes: File | New... offers me a new virtual machine. What you put in the box isn't really relevant at the level of the question.

    My original post was clear - I think it's just taken me to push back on the poor responses to get you to read it properly. And you still haven't given a decent answer (depends of course on whether IO bus & Disk utilisation is significant in a given scenario), instead once more choosing to talk down to me.

    Guys: if you don't know the answer then it's fine to say so, but to make me wrong over it is very bad form indeed. I've seen so many posts from the Support/Experts on this forum that don't address the question, that insult the intelligence of the poster, or even in the case of Abdul above - try to sell me Parallels. As I said before - it's a superb product damaged by the experience of costly upgrades, in-app 3rd party selling (Kaspersky and Acronis), and poor support.
     
  9. Specimen

    Specimen Product Expert

    Messages:
    3,242
    I have made much much more effort than you have in trying to understand and to explain. And in the process I've tried to clear up some possible misunderstandings. I'm out of this thread, my last recommendation (very patronising, I'm sure!) is that you just read everything in this thread carefully, diverting all that negative energy to actually reading, including what you wrote.
     
    Last edited: Oct 26, 2014
  10. Andrew@Parallels

    Andrew@Parallels Parallels Team

    Messages:
    633
    Thread is closed to avoid further off-topic discussions.
     
Thread Status:
Not open for further replies.

Share This Page