The generic tuning parameters listed here are available in the /sys/block/sdX/queue/ directory.

The following listed tuning parameters are separate from I/O scheduler tuning, and are applicable to all I/O schedulers:

add_random

Some I/O events contribute to the entropy pool for the /dev/random. This parameter can be set to 0 if the overhead of these contributions become measurable.

iostats

By default, iostats is enabled and the default value is 1. Setting iostats to 0 disables gathering of I/O statistics for the device. This removes a small amount of overhead with the I/O path.

Setting iostats to 0 might improve performance for high performance devices, such as certain Non-volatile Memory Express (NVMe) storage devices. It is preferable to leave iostats enabled unless otherwise specified for the given storage model by the vendor.

If you disable iostats, the I/O statistics for the device are no longer present within the /proc/diskstats file. The content of /sys/diskstats file is the source of I/O information for monitoring I/O tools, such as sar or iostats. Therefore, if you disable the iostats parameter for a device, it is no longer present in the output of I/O monitoring tools.

max_sectors_kb

Specifies the maximum size of an I/O request in kilobytes. The default value is 512 KB. The minimum value for this parameter is determined by the logical block size of the storage device. The maximum value for this parameter is determined by the value of the max_hw_sectors_kb.

max_sectors_kb must always be a multiple of the optimal I/O size and the internal erase block size. Use a value of logical_block_size for either parameter if they are zero or not specified by the storage device.

nomerges

Most workloads benefit from request merging. However, disabling merges can be useful for debugging purposes. By default, the nomerges parameter is set to 0, which enables merging. To disable simple one-hit merging, set nomerges to 1. and to disable all types of merging, set nomerges to 2.

nr_requests

It is the maximum allowed number of the queued I/O. If the current I/O scheduler is none, this number can only be reduced; otherwise the number can be increased or reduced.

optimal_io_size

Some storage devices report an optimal I/O size through this parameter. If this value is reported, applications issue I/O aligned to and in multiples of the optimal I/O size wherever possible.

read_ahead_kb

Defines the maximum number of kilobytes that the operating system may read ahead during a sequential read operation. As a result, the necessary information is already present within the kernel page cache for the next sequential read. This improves read I/O performance.

Device mappers often benefit from a high read_ahead_kb value. 128 KB for each device to be mapped is a good starting point. Increasing the read_ahead_kb value up to request queue’s max_sectors_kb of the disk might improve performance where sequential reading of large files occur.

rotational

Some solid-state disks do not correctly advertise their solid-state status, and are mounted as traditional rotational disks. Manually set the rotational value to 0 to disable unnecessary seek-reducing logic in the scheduler.

rq_affinity

The default value of the rq_affinity is 1. It completes the I/O operations on one CPU core, which is in the same CPU group of the issued CPU core. To perform completions only on the processor that issued the I/O request, set the rq_affinity to 2. To disable the mentioned two abilities, set it to 0.

scheduler

To set the scheduler or scheduler preference order for a storage device, edit the /sys/block/devname/queue/scheduler. Replace devname with the device name you want to configure.