VMware Player vs. VirtualBox: performance comparison

If you are using a virtualization hypervisor, one of your main concerns will be its performance, or in another word, its virtualization overhead. How much overhead is introduced by the virtualization layer will determine the raw performance of guest virtual machines (VMs) running on a hypervisor.

Continuing from the feature comparison between VMware Player and VirtualBox, I will present performance comparison between VMware Player and VirtualBox.

To compare virtualization overhead of two virtualization software packages, I created one VM on VMware Player, and another VM on VirtualBox, both of which have exactly the same specifications. Then I ran various PTS tests to obtain performance benchmarks from both VMs. Since two guest VMs have exactly the same hardware specification, the lower the virtualization overhead is, the better performance a VM will get.

I conducted the performance experiments on one VM at a time, so that there is no resource contention between the two VMs. That is, I powered on VMware Player VM, completed a series of tests on it, and powered off the VM. Next, I powered on VirtualBox VM, repeated the same set of tests with it, and powered it off.

Host Machine Specification

  • CPU: Intel(R) Core(TM) i7-3612QM CPU @ 2.10GHz
  • Memory: 8GB Shared Dual Channel DDR3 Memory
  • Disk: 1TB 5400 RPM SATA Hard Drive
  • VMware Player: 5.0.1 build-894247
  • VirtualBox: 4.2.16 r86992

Guest VM Specification (for both VMware Player and VirtualBox)

  • Processor count: 1
  • Memory: 1GB
  • Hard disk: 10GB
  • Filesystem: EXT4
  • Operating System: Ubuntu 13.04 (64-bit)
  • Linux Kernel: 3.8.0-19-generic (x86_64)

Guest VM Installation

  • VMware Player VM: VMware Tools
  • VirtualBox VM: Guest Additions

I did performance tests in three categories: (1) disk, (2) processor and (3) memory. For each category, I randomly chose several tests among available ones from Phoronix Test Suite.

Disk Performance Comparison

  • BlogBench: Final score. More is better. VMware Player won!

  • Dbench: MB/s. More is better. VMware Player won!

  • IOzone (Read): 64Kb record, 512MB file. MB/s. More is better. VirtualBox won!

  • IOzone (Write): 64Kb record, 512MB file. MB/s. More is better. VirtualBox won!

    CPU Performance Comparison

  • Gzip Compression: 2GB file. Seconds. Less is better. VMware Player won!

  • FFmpeg: Convert H.264 HD to NTSC DV. Seconds. Less is better. VMware Player won!

  • OpenSSL: RSA 4096-bit performance. Signs per seconds. More is better. VirtualBox won!

    Memory Performance Comparison

  • RAMspeed (Integer): MB/s. More is better. VMware Player won!

  • RAMspeed (Floating Point): MB/s. More is better. VMware Player won!

    Conclusion

    According to the test results, it appears that VMware Player has a slight edge over VirtualBox in terms of CPU and memory. As far as disk performance is concerned, it is not conclusive which one is better.

    It may still be hard to generalize the findings reported here, since this is a case of one particular hardware setting with a limited set of tests. So take it with a grain of salt.

    If you want to share your experience with VMware Player or VirtualBox in terms of their performance, feel free to share it in the comment section.

    Subscribe to Xmodulo

    Do you want to receive Linux FAQs, detailed tutorials and tips published at Xmodulo? Enter your email address below, and we will deliver our Linux posts straight to your email box, for free. Delivery powered by Google Feedburner.

  • 21 thoughts on “VMware Player vs. VirtualBox: performance comparison

      • Thanks for catching that. Yes, both VMware Tools and VirtualBox guest addition were installed on the guest VMs. I updated the article to clarify.

    1. Do these tests include the optional kernel modules for each setup (virtualbox guest modules and vmware tools)?

    2. I would have been more interested in seeing this test done with a Windows guest, considering that when Linux users are virtualizing on the desktop, it's usually for the purpose of running some crappy Windows application that they've been forced to use.

      In any case, it looks like the performance of these two tools is "in the neighborhood" of being identical. I'm partial to VMware but on my laptop I'm running VirtualBox. Why? Because it's in the Debian repositories, and upgrades itself along with the rest of my system.

      • Recent versions of Wine 1.6 are so good that this may not be the case any more.

        Every Windows application I met since Wine 1.6 release runs fine, without the need to run the full OS in Virtualbox. Maybe we can say goodbye to virtualization in this use case (it's reallly useful in other fields). No more wasted HD space and RAM to run a crappy OS! :-)

        • Actually gaming in virtual machines is quite popular. Xen pass-through enables the use of bare metal hardware for a VM. It is a Linux kernel module and header. VMware Workstation soars in this category meanwhile the differences become more and more magnified. Virtualbox just chokes on Xen. At this point it is like running the virtualized OS on bare metal meanwhile still sandboxing the main OS from the virtualized one. Take a snapshot of Windows and if it gets a virus or gobbled up, simply re-load said snapshot. It takes like 30 seconds whereas installing windows takes a considerable amount of time.

          USB support is amazing too... I just have no hope for V-box anymore...

          I am not sure if VMware Player can do it as I own Workstation 10 and it runs fantastically. The extra step of booting everything up makes it well worth the time it takes by far.

          Don't get me wrong but it seems a lot of FOSS seems doomed except for the exceptionally rare like the Linux kernel or Gnome 3.12... Firefox is incredible too. I feel it is quintessential to also note that gaming in a virtual environment takes the fear out of being hacked for companies.

    3. Why are you so miserly, making processors available to the guests? At least in VirtualBox, the number of processors defines the maximum number of virtual cores usable by the guest. I always provide 4 (of totally 8), even running more than one guest. The benchmarks utilize multiple threads. Could providing more processors impact the results?

    4. I am able to capture h.264 video with vmware (with spec USB 3.0). Virtual box seems can only enumerate USB 1-1 devices which is far from being useful.

    5. Nice article. Not just some meaningless screenshots about what the products look like, some real meat here.
      Thanks

    6. Why don't they update their VM players to include USB2.0. It's a pain to go shopping for an extension pack, and even a bigger pain to download it. Too many gotyas.

    7. You didnt say what was the host OS. Maybe Windows host vs Linux host would be interesting comparision.

    8. On a couple of the disk tests, it seems the speed of disk itself is the limiting factor. I'd love to see the CPU usage when doing those operations. that would probably be the easiest way to get results that would correlate to real-world I/O performance without having to blow $500 on a hyper-fast SSD.

      Keep up the good work!

    Leave a comment

    Your email address will not be published. Required fields are marked *

    Current day month ye@r *