Shulou(Shulou.com)06/01 Report--

This article is about how to use cgroups to manage the process disk io in Linux. The editor thinks it is very practical, so share it with you as a reference and follow the editor to have a look.

Linux's cgroups can also limit and monitor the disk io of the process. This function is implemented through the blkio subsystem.

There are many things in the blkio subsystem. However, most of them are read-only status reports, and the only parameters that can be written are the following:

The code is as follows:

Blkio.throttle.read_bps_device

Blkio.throttle.read_iops_device

Blkio.throttle.write_bps_device

Blkio.throttle.write_iops_device

Blkio.weight

Blkio.weight_device

These are used to control the process of the disk io. It is clearly divided into two categories, of which those with "throttle", as the name implies, are throttle valves that limit flow to a certain value. And "weight" is to assign the weight of io.

The four parameters of "throttle" can tell what they are for by looking at the name. Use blkio.throttle.read_bps_device to limit the number of bytes that can be read per second. Run some io first.

The code is as follows:

Dd if=/dev/sda of=/dev/null &

[1] 2750

Look at the current io with iotop

The code is as follows:

TID PRIO USER DISK READ DISK WRITE SWAPIN IO > COMMAND

2750 be/4 root 66.76 MUnix s 0.00 BUnip s 0.00% 68.53% dd if=/dev/sda of=/dev/null

...

Then modify the resource limit to add the process to the control group

The code is as follows:

Echo '8:0 1048576' > / sys/fs/cgroup/blkio/foo/blkio.throttle.read_bps_device

Echo 2750 > / sys/fs/cgroup/blkio/foo/tasks

8:0 here is the main equipment number and secondary equipment number of the corresponding block equipment. It can be viewed by the ls-l device file name. Such as

The code is as follows:

# ls-l / dev/sda

Brw-rw----. 1 root disk 8, 0 Oct 24 11:27 / dev/sda

The 8,0 here is the corresponding device number. Therefore, cgroups can impose different restrictions on different devices. Then let's see the effect.

The code is as follows:

TID PRIO USER DISK READ DISK WRITE SWAPIN IO > COMMAND

2750 be/4 root 989.17 Khand s 0.00 Bhand s 0.00% 96.22% dd if=/dev/sda of=/dev/null

...

As you can see, the read per second of the process is immediately reduced to around 1MB. To lift the restriction, write such as "8:0 0" to the file

It is important to note, however, that this approach is not useful for large amounts of io generated less than in the sampling interval. For example, even if there is a peak of 100m writes per second in 1s, it will not be limited.

And look at blkio.weight. The throttle and weight methods of blkio are somewhat similar to the quota and shares of the cpu subsystem, one is absolute limit, the other is relative limit, and can make full use of resources when not busy, with weights ranging from 10 to 1000.

Testing weights is a bit more troublesome. Because it is not an absolute limit, it is affected by the file system cache. If testing in a virtual machine, turn off the cache of the virtual machine such as the VirtualBox I used on the host. If you want to test the effect of reading io, you can use dd to make two large files / tmp/file_1,/tmp/file_2 with several gigabytes. Then set two weights

The code is as follows:

# echo 500 > / sys/fs/cgroup/blkio/foo/blkio.weight

# echo 100 > / sys/fs/cgroup/blkio/bar/blkio.weight

Clear the file system cache before testing so as not to interfere with the test results

The code is as follows:

Sync

Echo 3 > / proc/sys/vm/drop_caches

Dd was used to produce io test results in these two control groups.

The code is as follows:

# cgexec-g "blkio:foo" dd if=/tmp/file_1 of=/dev/null &

[1] 1838

# cgexec-g "blkio:bar" dd if=/tmp/file_2 of=/dev/null &

[2] 1839

Or use iotop to see the effect.

The code is as follows:

TID PRIO USER DISK READ DISK WRITE SWAPIN IO > COMMAND

1839 be/4 root 48.14 Mhand s 0.00 Bhand s 0.00% 99.21% dd if=/tmp/file_2 of=/dev/null

1838 be/4 root 223.59 M Union s 0.00 Bhand s 0.00% 16.44% dd if=/tmp/file_1 of=/dev/null

Although the number of bytes read by the two processes per second changes, the general trend remains at around 1:5, in line with the set weight ratio. Blkio.weight_device is sub-device. When writing, add the device number in front of it.

There are many statistical items in the blkio subsystem.

Blkio.time

Io access time for each device (in milliseconds)

Blkio.sectors

Number of sectors swapped in or out of each device

Blkio.io_serviced

Various types of io operands performed in each device, including read, write, sync, async, and total

Blkio.io_service_bytes

The number of bytes of each type of io switch in or out of each device

Blkio.io_service_time

Each type of io time performed in each device, in microseconds

Blkio.io_wait_time

The waiting time of each type of io in the queue in each device

Blkio.io_merged

The number of io requests merged for each type in each device

Blkio.io_queued

The number of io requests currently in queue for each type of device

Use these statistical items to better count and monitor the io of the process.

Use

The code is as follows:

Echo 1 > blkio.reset_stats

You can zero all statistical items.

Thank you for reading! This is the end of this article on "how to use cgroups management process disk io in Linux". I hope the above content can be of some help to you, so that you can learn more knowledge. if you think the article is good, you can share it out for more people to see!

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.