Chapter 11. Setting the disk scheduler
The disk scheduler is responsible for ordering the I/O requests submitted to a storage device.
You can configure the scheduler in several different ways:
- Set the scheduler using TuneD, as described in Setting the disk scheduler using TuneD
-
Set the scheduler using
udev
, as described in Setting the disk scheduler using udev rules - Temporarily change the scheduler on a running system, as described in Temporarily setting a scheduler for a specific disk
In Red Hat Enterprise Linux 9, block devices support only multi-queue scheduling. This enables the block layer performance to scale well with fast solid-state drives (SSDs) and multi-core systems.
The traditional, single-queue schedulers, which were available in Red Hat Enterprise Linux 7 and earlier versions, have been removed.
11.1. Available disk schedulers
The following multi-queue disk schedulers are supported in Red Hat Enterprise Linux 9:
none
- Implements a first-in first-out (FIFO) scheduling algorithm. It merges requests at the generic block layer through a simple last-hit cache.
mq-deadline
Attempts to provide a guaranteed latency for requests from the point at which requests reach the scheduler.
The
mq-deadline
scheduler sorts queued I/O requests into a read or write batch and then schedules them for execution in increasing logical block addressing (LBA) order. By default, read batches take precedence over write batches, because applications are more likely to block on read I/O operations. Aftermq-deadline
processes a batch, it checks how long write operations have been starved of processor time and schedules the next read or write batch as appropriate.This scheduler is suitable for most use cases, but particularly those in which the write operations are mostly asynchronous.
bfq
Targets desktop systems and interactive tasks.
The
bfq
scheduler ensures that a single application is never using all of the bandwidth. In effect, the storage device is always as responsive as if it was idle. In its default configuration,bfq
focuses on delivering the lowest latency rather than achieving the maximum throughput.bfq
is based oncfq
code. It does not grant the disk to each process for a fixed time slice but assigns a budget measured in number of sectors to the process.This scheduler is suitable while copying large files and the system does not become unresponsive in this case.
kyber
The scheduler tunes itself to achieve a latency goal by calculating the latencies of every I/O request submitted to the block I/O layer. You can configure the target latencies for read, in the case of cache-misses, and synchronous write requests.
This scheduler is suitable for fast devices, for example NVMe, SSD, or other low latency devices.
11.2. Different disk schedulers for different use cases
Depending on the task that your system performs, the following disk schedulers are recommended as a baseline prior to any analysis and tuning tasks:
Use case | Disk scheduler |
---|---|
Traditional HDD with a SCSI interface |
Use |
High-performance SSD or a CPU-bound system with fast storage |
Use |
Desktop or interactive tasks |
Use |
Virtual guest |
Use |
11.3. The default disk scheduler
Block devices use the default disk scheduler unless you specify another scheduler.
For non-volatile Memory Express (NVMe)
block devices specifically, the default scheduler is none
and Red Hat recommends not changing this.
The kernel selects a default disk scheduler based on the type of device. The automatically selected scheduler is typically the optimal setting. If you require a different scheduler, Red Hat recommends to use udev
rules or the TuneD application to configure it. Match the selected devices and switch the scheduler only for those devices.
11.4. Determining the active disk scheduler
This procedure determines which disk scheduler is currently active on a given block device.
Procedure
Read the content of the
/sys/block/device/queue/scheduler
file:# cat /sys/block/device/queue/scheduler [mq-deadline] kyber bfq none
In the file name, replace device with the block device name, for example
sdc
.The active scheduler is listed in square brackets (
[ ]
).
11.5. Setting the disk scheduler using TuneD
This procedure creates and enables a TuneD profile that sets a given disk scheduler for selected block devices. The setting persists across system reboots.
In the following commands and configuration, replace:
-
device with the name of the block device, for example
sdf
-
selected-scheduler with the disk scheduler that you want to set for the device, for example
bfq
Prerequisites
-
The
TuneD
service is installed and enabled. For details, see Installing and enabling TuneD.
Procedure
Optional: Select an existing TuneD profile on which your profile will be based. For a list of available profiles, see TuneD profiles distributed with RHEL.
To see which profile is currently active, use:
$ tuned-adm active
Create a new directory to hold your TuneD profile:
# mkdir /etc/tuned/my-profile
Find the system unique identifier of the selected block device:
$ udevadm info --query=property --name=/dev/device | grep -E '(WWN|SERIAL)' ID_WWN=0x5002538d00000000_ ID_SERIAL=Generic-_SD_MMC_20120501030900000-0:0 ID_SERIAL_SHORT=20120501030900000
NoteThe command in the this example will return all values identified as a World Wide Name (WWN) or serial number associated with the specified block device. Although it is preferred to use a WWN, the WWN is not always available for a given device and any values returned by the example command are acceptable to use as the device system unique ID.
Create the
/etc/tuned/my-profile/tuned.conf
configuration file. In the file, set the following options:Optional: Include an existing profile:
[main] include=existing-profile
Set the selected disk scheduler for the device that matches the WWN identifier:
[disk] devices_udev_regex=IDNAME=device system unique id elevator=selected-scheduler
Here:
-
Replace IDNAME with the name of the identifier being used (for example,
ID_WWN
). Replace device system unique id with the value of the chosen identifier (for example,
0x5002538d00000000
).To match multiple devices in the
devices_udev_regex
option, enclose the identifiers in parentheses and separate them with vertical bars:devices_udev_regex=(ID_WWN=0x5002538d00000000)|(ID_WWN=0x1234567800000000)
-
Replace IDNAME with the name of the identifier being used (for example,
Enable your profile:
# tuned-adm profile my-profile
Verification
Verify that the TuneD profile is active and applied:
$ tuned-adm active Current active profile: my-profile
$ tuned-adm verify Verification succeeded, current system settings match the preset profile. See TuneD log file ('/var/log/tuned/tuned.log') for details.
Read the contents of the
/sys/block/device/queue/scheduler
file:# cat /sys/block/device/queue/scheduler [mq-deadline] kyber bfq none
In the file name, replace device with the block device name, for example
sdc
.The active scheduler is listed in square brackets (
[]
).
Additional resources
11.6. Setting the disk scheduler using udev rules
This procedure sets a given disk scheduler for specific block devices using udev
rules. The setting persists across system reboots.
In the following commands and configuration, replace:
-
device with the name of the block device, for example
sdf
-
selected-scheduler with the disk scheduler that you want to set for the device, for example
bfq
Procedure
Find the system unique identifier of the block device:
$ udevadm info --name=/dev/device | grep -E '(WWN|SERIAL)' E: ID_WWN=0x5002538d00000000 E: ID_SERIAL=Generic-_SD_MMC_20120501030900000-0:0 E: ID_SERIAL_SHORT=20120501030900000
NoteThe command in the this example will return all values identified as a World Wide Name (WWN) or serial number associated with the specified block device. Although it is preferred to use a WWN, the WWN is not always available for a given device and any values returned by the example command are acceptable to use as the device system unique ID.
Configure the
udev
rule. Create the/etc/udev/rules.d/99-scheduler.rules
file with the following content:ACTION=="add|change", SUBSYSTEM=="block", ENV{IDNAME}=="device system unique id", ATTR{queue/scheduler}="selected-scheduler"
Here:
-
Replace IDNAME with the name of the identifier being used (for example,
ID_WWN
). -
Replace device system unique id with the value of the chosen identifier (for example,
0x5002538d00000000
).
-
Replace IDNAME with the name of the identifier being used (for example,
Reload
udev
rules:# udevadm control --reload-rules
Apply the scheduler configuration:
# udevadm trigger --type=devices --action=change
Verification
Verify the active scheduler:
# cat /sys/block/device/queue/scheduler
11.7. Temporarily setting a scheduler for a specific disk
This procedure sets a given disk scheduler for specific block devices. The setting does not persist across system reboots.
Procedure
Write the name of the selected scheduler to the
/sys/block/device/queue/scheduler
file:# echo selected-scheduler > /sys/block/device/queue/scheduler
In the file name, replace device with the block device name, for example
sdc
.
Verification
Verify that the scheduler is active on the device:
# cat /sys/block/device/queue/scheduler