KVM disk performance: IDE vs VirtIO

If you use QEMU-KVM (or virt-manager GUI) for running your virtual machines, you can specify a disk driver to be used for accessing the machine’s disk image. By default IDE is selected, but VirtIO is a very popular choice as well. How do they compare in performance?

I used Fedora 17 as a host, Fedora 18 as a guest system. I used bonnie++ for testing. I ran

$ bonnie++ -s 3g

twice for each driver. The disk images were in the form of raw files. The results are:

IDE round 1:

Version  1.96       ------Sequential Output------ --Sequential Input- --Random-
Concurrency   1     -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine        Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP  /sec %CP
localhost        3G    76  97 24645  25 15658  17  1229  97 277235  56 717.8  65
Latency               196ms    3233ms     804ms   10194us    7619us   90962us
Version  1.96       ------Sequential Create------ --------Random Create--------
localhost           -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
              files  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP
                 16  4738  72 +++++ +++  7411  62  4893  74 +++++ +++  7853  59
Latency              1019us    6430us    7828us     693us     174us     824us

IDE round 2:

Version  1.96       ------Sequential Output------ --Sequential Input- --Random-
Concurrency   1     -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine        Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP  /sec %CP
localhost        3G    68  87 24893  26 15971  17  1063  85 282355  56 730.7  66
Latency               260ms     679ms     734ms   43305us    8152us   85837us
Version  1.96       ------Sequential Create------ --------Random Create--------
localhost           -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
              files  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP
                 16  4740  72 +++++ +++  7488  65  4833  73 +++++ +++  7757  65
Latency              2480us    6227us    7094us     745us     251us     760us

VirtIO round 1:

Version  1.96       ------Sequential Output------ --Sequential Input- --Random-
Concurrency   1     -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine        Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP  /sec %CP
localhost        3G    77  98 56770  38 32139  19  1222  98 392441  49 937.4  53
Latency               167ms     512ms     666ms    9544us    7309us   85529us
Version  1.96       ------Sequential Create------ --------Random Create--------
localhost           -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
              files  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP
                 16  5303  81 +++++ +++  8301  74  5398  81 +++++ +++  8851  68
Latency              3725us    6628us    7241us     738us     970us     795us

VirtIO round 2:

Version  1.96       ------Sequential Output------ --Sequential Input- --Random-
Concurrency   1     -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine        Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP  /sec %CP
localhost        3G    77  97 57202  35 32640  19  1230  98 395489  48 972.4  50
Latency               241ms    1478ms     684ms    9326us   81461us   84019us
Version  1.96       ------Sequential Create------ --------Random Create--------
localhost           -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
              files  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP
                 16  4753  74 +++++ +++  8434  71  5473  82 +++++ +++  8987  69
Latency              8461us    5904us    7358us     706us      72us     779us

The results (see description) are pretty clear. VirtIO is substantially faster for disk I/O operations. If you don’t use it, you should definitely consider it. If you run a lot of installation tests or other heavy disk operations like I do, it can be a pretty significant difference in performance.

Flattr this

Advertisements
KVM disk performance: IDE vs VirtIO

5 thoughts on “KVM disk performance: IDE vs VirtIO

  1. lzap says:

    “By default IDE is used” – afaik virtio is used too, if you select appropriate os in the libvirt-manager. Like RHEL 6+ or Fedora 15+. Nice testing!

    1. You’re right, if I create a new Linux machine, it uses virtio. But if I edit an existing Linux machine and choose Add new storage, IDE is the default. Bad, bad virt-manager!

  2. ted says:

    not a surprise… but on some guest we have approximatively the same performance inside the VMs, but not using virtio cost a lot of cpu on the hypervisor.

Leave a Reply (Markdown syntax supported)

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s