Use DD to Quickly Benchmark Your CPU

  • 0

Use DD to Quickly Benchmark Your CPU

Category : How-to

Get Social!

Let me start by saying… this is a quick and dirty method and shouldn’t be used for precise comparisons of CPU performance. That said, it’s perfectly adequate for approximating CPU performance, especially on Cloud hosted VPSs to ensure you’re getting the horsepower that you’re being promised.

The idea here is to force your machine to perform tasks that will be computationally expensive to force your CPU to work at 100 percent and become the bottleneck for the task (rather than disk I/ O, etc.). This task will then be timed. The shorter times will generally represent faster CPU’s and longer results would indicate a slower CPU.

CPU benchmark

The md5sum command is a tool that creates an MD5 hash of some data. We can generate some data on the fly with dd and pipe it into the md5sum tool to create a computationally expensive task. We’ll limit the data to hash and time the length of time it takes to create the hash.

Run the below to start the test. If your result completes in under 2 seconds then increase the count=1k value to a higher value, for example count=10k.

You’ll get an output similar to the below output.

There are a couple of items that you’re interested in here, and the rest can be ignored.

  • 2.38909 s is the time it took in seconds for the operation to complete. This is the number to use for comparison with other machines – lower is better.
  • 449 MB/s is the speed that the data was fabricated and push into the md5sum tool to be hashed – higher is better.

 

CPU details with cpuinfo

Linux has various nuggets of information about your system available in the proc directory on a linux root partition. You can cat various files, such as /proc/cpuinfo, to see system specifications and metrics.

The output will look similar to the below output that shows a Xeon CPU running at a clock speed of 2.50GHz.

 


  • 1

Use A File As A Linux Block Device

Get Social!

Just like when creating a SWAP file, you can create a file on a disk and present it as a block device. The block device would have a maximum file size of the backing file, and (as long as it’s not in use) be moved around like a normal file. For example, I could create a 1GB file on the filesystem and make Linux treat the file as a disk mounted in /dev/. And guess what – that’s what we’re going to do.

Create a file and filesystem to use as a block device

First off, use dd to create a 1GB file on an existing disk that we’ll use for our storage device:

Then ‘format’ the file to give it the structure of a filesystem. For this example we’re going to use ext4 but you could choose any filesystem that meets your needs.

You’ll be promoted with Proceed anyway?. Type y and press return to proceed with the process.

Mounting a loop device

Before mounting the file we need to check that there is a free /dev/loopX loopback device that we can use to represent our new block device.

Run the below command, and if there is any output then check if it’s one of your loop devices, which will more than likely reference /dev/loop as the mounted device. If you do have a reference to our loop device then see the below section on Unmounting a loop device, or choose a number higher than the highest listed loop device, for example: usually there are several loop devices, starting with loop0 and going up in value to loop1loop2, and so on.

Once you have the file that you’d like to mount and a free loop device then you can go ahead and mount the file as a block device. You have two options:

  1. Mount the file as a block device only
  2. Mount the file as a block device and mount the filesystem of it on a local mount point (eg. /mnt/mymountpoint).

For option 1; to only mount the file as a device in /dev/, run the below command and change /root/diskimage to the path of the file you’d like to mount. loop0 can also be incremented as explained above.

If you’d like this to be remounted after a machine reboot then add the above line to the rc.local file.

And add:

 

For option 2; to mount the file and the filesystem on it, use the mount command. You must have already created the mount point locally before running the command, as you would when mounting a disk or NFS share.

Then run the mount command and specify the loop device, the path of the file and the path to mount the filesystem on:

To check the file has been mounted you can use the df command:

Unmounting a loop device

If you’ve mounted the filesystem on the block device using the mount command then make sure it’s unmounted before proceeding.

To then free the loop0 device (or which ever loop device you’ve used) you’ll need the losetup command with the d switch.

 


  • 1

dd Cheat Sheet

Get Social!

dd is one of the most versatile IO tools available for Linux. It’s used in a variety of ways ranging from Disk Benchmarking through to creating SWAP files and copying downloaded disk images to physical disks.

dd takes the following common switches:

  • if is the input file name and location.
  • of is the name and location of the output file.
  • bs is the block size that will be used to read and/ or write the file. Increasing this can help with performance  or dictate how much data will be read or written.
  • count is the number of blocks that will be used.
  • seek is the number of blocks on the output file that will be skipped before writing any data.
  • skip is the number of blocks that will be skipped on the input file before starting to read data.
  • conv is a comma separated list of additional parameters that can be used. See the man dd for more information.

The below headings will list a few example uses of dd in a typical Linux environment.

Backup disk partition with dd

You can use dd to copy an entire disk partition to a virtual disk file. This can be useful for creating a backup or to clone the disk to another machine.

You can use this method to read a CD-ROM, USB drive or Flash disk to a file in the same way – just make sure the device is inserted and point the if= part of the dd command to the relevant /dev/ device.

You could also compress the image as part of the process with gzip.

Restore disk partition with dd

Similar to the above command, you can use dd to replace a disk’s partition with a virtual disk file.

If you compressed the image then you can decompress it first all in one go:

Create a fixed size file with dd

You can create a fixed size file with DD that will be created in the location you specify.

This will create a file in /root/test of 1024 bytes in size. Increase either bs or count to change the size of the file. The resulting size will be bs count. You can also use shorhand sizes such as K, M and G with bs, for example bs=1G,

Create a SWAP file with dd

dd can be used to create a SWAP file that can be used as a SWAP device by your computer. This is often needed with smaller instances on Cloud providers such as AWS.

The starting point is the same as the above command to create a file with the size that you’d like to use for swap. See my other blog post for more info.

Split a file with dd

dd can be used to read just part of a file, given offset and length coordinates. The below example will skip the first 100 bytes of the file and output the proceeding 10 bytes (byte 101 – 111).

You could repeat this process to split a large file into multiple smaller files, to be able to email it for example.

Merge multiple files with dd

You can merge multiple files into a single file with dd. Following on from the above split example, the below will rejoin the 3 file parts into a single file.

Convert text to lower case with dd

You can use the conv switch with dd to transform ascii text from upper case to lower case and visa-versa. Using lcase and ucase in the conv switch will instruct dd to convert the text as it’s written.

The below example will convert all characters in the filetoconvert.txt. file to lower case.

 


  • 12

Benchmark disk IO with DD and Bonnie++

Get Social!

Benchmarking disk or file system IO performance can be tricky at best. The problem is that modern file systems leverage various techniques to ensure that the best performance is achieved such as caching files in RAM. This means that unless you circumvent the disk cache, your reported speeds will be reporting how quickly the files can be read from memory.

In this example, I’ll cover benchmarking a Linux file system using two methods; dd for the easy route, and bonnie++ for a more comprehensive test.

dd

Write

You can use dd to create a large file as quickly as possible to see how long it takes. It’s a very basic test and not very customisable however it will give you a sense of the performance of the file system. You must make sure this file is larger than the amount of RAM you have on your system to avoid the whole file being cached in memory.

It’s usually installed out-of-the-box with most Linux file systems which makes it an ideal tool in locked-down environments or environments where it’s tricky to get packages installed onto. Use the below command substituting [PATH] with the filesystem path to test, [BLOCK_SIZE] with the block size and [LOOPS] for the amount of blocks to write.

A break down of the command is as follows:

  • time – times the overall process from start to finish
  • of= this is the path which you would like to test. The path must be read/ writable.
  • bs= is the block size to use. If you have a specific load which you are testing for, make this value mirror the write size which you would expect.
  • sync – forces the process to write the entire file to disk before completing. Note, that dd will return before completing but the time command will not, therefore the time output will include the sync to disk.

The below example uses a 4K block size and loops 2000000 times. The resulting write size will be around 7.6GB.

Now, let’s do the math. dd tells us how many bytes were written, and the time command tells us how long it took – use the real output at the bottom of the output. Use the formula BYTES / SECONDS. For these larger tests, convert bytes to KB or MB to make more sensible numbers.

(8192000000 / 1024 / 1024) / ((2 * 60) + 41.618)

Bytes converted to MB / (2 minutes + 41.618 seconds)

This gives us an average of 48.34 megabytes per second over the duration of the test.

Read

We can also use dd to test the read speed of a disk by reading the file we created and timing the process. Before we do that, we need to flush the file cache by writing another file which is about the size of the RAM installed on the test system. If we don’t do this, the file we just created will be partially in RAM and therefore the read test will not be completely read from disk.

Create a file using dd which is about the same size as the RAM installed on the system. The below assumes 2GB of RAM is installed. You can check how much RAM is installed with free.

Now for the read test of our original file.

And process the time result the same was as when writing.

Bonnie++

Bonnie++ is a small utility with the purpose of benchmarking file system IO performance. It’s commonly available in Linux repositories or available from source from the home page.

On Debian/ Ubuntu based systems, use the apt-get command.

Just like with DD, we need to minimise the effect of file caching and therefore the tests should be performed on datasets larger than the amount of RAM you have on the test system. Some people suggest that you should use datasets up to 20 times the amount of RAM, others suggest twice the amount of RAM. Whichever you use, always use the same dataset size for all tests performed to ensure the results are comparable.

There are many commands which can be used with bonnie++, too many to cover here so let’s look at some of the common ones.

  • -d – is used to specify the file system directory to use to benchmark.
  • -u – is used to run a a particular user. This is best used if you run the program as root. This is the UID or the name.
  • -g – is used to run as a particular group. This is the GID or the name.
  • -r – is used to specify the amount of RAM in MB the system has installed. This is total RAM, and not free RAM. Use free -m to find out how much RAM is on your system.
  • -b – removes write buffering and performs a sync at the end of each bonnie++ operation.
  • -s – specifies the dataset size to use for the IO test in MB.
  • -n – is the number of files to use for the create files test.
  • -m – this adds a label to the output so that you can understand what the test was at a later date.
  • -x – is used to repeat the tests n times. Change n to the number of how many times to run the tests.

bonnie++ performs multiple tests, depending on the arguments used, and does not display much until the tests are complete. When the tests complete, two outputs are visible. The bottom line is not readable (unless you really know what you are doing) however above that is a table based output of the results of the tests performed.

Let’s start with a basic test, telling bonnie++ where to test and how much RAM is installed, 2GB in this example. bonnie++ will then use a dataset twice the size of the RAM for tests. As I am running as root, I am specifying a user name.

bonnie++ will take a few minutes, depending on the speed of your disks and return with something similar to the output below.

The output shows quite a few statistics, but it’s actually quite straight forward once you understand the format. First, discard the bottom line (or three lines in the above output) as this is the results separated by a comma. Some scripts and graphing applications understand these results but it’s not so easy for humans. The top few lines are just the tests which bonnie++ performs and again, can be discarded.

Of cause, all the output of bonnie++ is useful in some context however we are just going to concentrate on random read/ write, reading a block and writing a block. This boils down to this section:

The above output is not the easiest output to understand due to the character spacing but you should be able to follow it, just. The below points are what we are interested in, for this example, and should give you a basic understanding of what to look for and why.

  • ubuntu is the machine name. If you specified -m some_test_info this would change to some_test_info.
  • 4GB is the total size of the dataset. As we didn’t specify -s, a default of RAM x 2 is used.
  • 17094 shows the speed in KB/s which the dataset was written. This, and the next three points are all sequential reads – that is reading more than one data block.
  • 15431 is the speed at which a file is read and then written and flushed to the disk.
  • 37881 is the speed the dataset is read.
  • 548.4 shows the number of blocks which bonnie++ can seek to per second.
  • Latency number correspond with the above operations – this is the full round-trip time it takes for bonnie++ to perform the operations.

Anything showing multiple +++ is because the test could not be ran with reasonable assurance on the results because they completed too quickly. Increase -n to use more files in the operation and see the results.

bonnie++ can do much more and, even out of the box, show much more but this will give you some basic figures to understand and compare. Remember, always perform tests on datasets larger than the RAM you have installed, multiple times over the day, to reduce the chance of other processes interfering with the results.