# Amazon EBS volumes and NVMe
Amazon EBS volumes are exposed as NVMe block devices on Amazon EC2 instances built on the [AWS Nitro System](https://docs.aws.amazon.com/ec2/latest/instancetypes/ec2-nitro-instances.html). To fully utilize the performance and capabilities of Amazon EBS volumes exposed as NVMe block devices, the EC2 instance must have the AWS NVMe driver installed. All current generation AWS Windows and Linux AMIs come with the AWS NVMe driver installed by default.
If you use an AMI that does not have the AWS NVMe driver, you can manually install it. For more information, see [AWS NVMe drivers](https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/aws-nvme-drivers.html) in the *Amazon EC2 User Guide*.
**Linux instances**
The device names are `/dev/nvme0n1`, `/dev/nvme1n1`, and so on. The device names that you specify in a block device mapping are renamed using NVMe device names (`/dev/nvme[0-26]n1`). The block device driver can assign NVMe device names in a different order than you specified for the volumes in the block device mapping.
**Windows instances**
When you attach a volume to your instance, you include a device name for the volume. This device name is used by Amazon EC2. The block device driver for the instance assigns the actual volume name when mounting the volume, and the name assigned can be different than the name that Amazon EC2 uses.
**Topics**
+ [Map Amazon EBS volumes to NVMe device names](identify-nvme-ebs-device.md)
+ [NVMe I/O operation timeout for Amazon EBS volumes](timeout-nvme-ebs-volumes.md)
+ [NVMe Abort command for Amazon EBS volumes](abort-command.md)
# Map Amazon EBS volumes to NVMe device names
EBS uses single-root I/O virtualization (SR-IOV) to provide volume attachments on Nitro-based instances using the NVMe specification. These devices rely on standard NVMe drivers on the operating system. These drivers typically discover attached devices during instance boot, and create device nodes based on the order in which the devices respond, not on how the devices are specified in the block device mapping.
## Linux instances
In Linux, NVMe device names follow the pattern `/dev/nvmen`, where is the enumeration order, and, for EBS, is 1. Occasionally, devices can respond to discovery in a different order in subsequent instance starts, which causes the device name to change. Additionally, the device name assigned by the block device driver can be different from the name specified in the block device mapping.
We recommend that you use stable identifiers for your EBS volumes within your instance, such as one of the following:
+ For Nitro-based instances, the block device mappings that are specified in the Amazon EC2 console when you are attaching an EBS volume or during `AttachVolume` or `RunInstances` API calls are captured in the vendor-specific data field of the NVMe controller identification. With Amazon Linux AMIs later than version 2017.09.01, we provide a `udev` rule that reads this data and creates a symbolic link to the block-device mapping.
+ The EBS volume ID and the mount point are stable between instance state changes. The NVMe device name can change depending on the order in which the devices respond during instance boot. We recommend using the EBS volume ID and the mount point for consistent device identification.
+ NVMe EBS volumes have the EBS volume ID set as the serial number in the device identification. Use the `lsblk -o +SERIAL` command to list the serial number.
+ The NVMe device name format can vary depending on whether the EBS volume was attached during or after the instance launch. NVMe device names for volumes attached after instance launch include the `/dev/` prefix, while NVMe device names for volumes attached during instance launch do not include the `/dev/` prefix.
+ For Amazon Linux or FreeBSD AMI, use the `sudo ebsnvme-id /dev/nvme0n1 -u` command for a consistent NVMe device name.
+ For other distributions, use the `sudo nvme id-ctrl -V /dev/nvme0n1` command to determine the NVMe device name. You might need to include the `--vendor-specific` command option.
+ When a device is formatted, a UUID is generated that persists for the life of the filesystem. A device label can be specified at the same time. For more information, see [Make an Amazon EBS volume available for use](ebs-using-volumes.md) and [ Boot from the wrong volume](https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/instance-booting-from-wrong-volume.html).
**Amazon Linux AMIs**
With Amazon Linux AMI 2017.09.01 or later (including Amazon Linux 2), you can run the **ebsnvme-id** command as follows to map the NVMe device name to a volume ID and device name:
The following example shows the command and output for a volume attached during instance launch. Note that the NVMe device name does not include the `/dev/` prefix.
```
[ec2-user ~]$ sudo /sbin/ebsnvme-id /dev/nvme0n1
Volume ID: vol-01324f611e2463981
sda
```
The following example shows the command and output for a volume attached after instance launch. Note that the NVMe device name includes the `/dev/` prefix.
```
[ec2-user ~]$ sudo /sbin/ebsnvme-id /dev/nvme1n1
Volume ID: vol-064784f1011136656
/dev/sdf
```
Amazon Linux also creates a symbolic link from the device name in the block device mapping (for example, `/dev/sdf`), to the NVMe device name.
**FreeBSD AMIs**
Starting with FreeBSD 12.2-RELEASE, you can run the **ebsnvme-id** command as shown above. Pass either the name of the NVMe device (for example, `nvme0`) or the disk device (for example, `nvd0` or `nda0`). FreeBSD also creates symbolic links to the disk devices (for example, `/dev/aws/disk/ebs/`*volume\$1id*).
**Other Linux AMIs**
With a kernel version of 4.2 or later, you can run the **nvme id-ctrl** command as follows to map an NVMe device to a volume ID. First, install the NVMe command line package, `nvme-cli`, using the package management tools for your Linux distribution. For download and installation instructions for other distributions, refer to the documentation specific to your distribution.
The following example gets the volume ID and NVMe device name for a volume that was attached during instance launch. Note that the NVMe device name does not include the `/dev/` prefix. The device name is available through the NVMe controller vendor-specific extension (bytes 384:4095 of the controller identification):
```
[ec2-user ~]$ sudo nvme id-ctrl -V /dev/nvme0n1
NVME Identify Controller:
vid : 0x1d0f
ssvid : 0x1d0f
sn : vol01234567890abcdef
mn : Amazon Elastic Block Store
...
0000: 2f 64 65 76 2f 73 64 6a 20 20 20 20 20 20 20 20 "sda..."
```
The following example gets the volume ID and NVMe device name for a volume that was attached after instance launch. Note that the NVMe device name includes the `/dev/` prefix.
```
[ec2-user ~]$ sudo nvme id-ctrl -V /dev/nvme1n1
NVME Identify Controller:
vid : 0x1d0f
ssvid : 0x1d0f
sn : volabcdef01234567890
mn : Amazon Elastic Block Store
...
0000: 2f 64 65 76 2f 73 64 6a 20 20 20 20 20 20 20 20 "/dev/sdf..."
```
The **lsblk** command lists available devices and their mount points (if applicable). This helps you determine the correct device name to use. In this example, `/dev/nvme0n1p1` is mounted as the root device and `/dev/nvme1n1` is attached but not mounted.
```
[ec2-user ~]$ lsblk
NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINT
nvme1n1 259:3 0 100G 0 disk
nvme0n1 259:0 0 8G 0 disk
nvme0n1p1 259:1 0 8G 0 part /
nvme0n1p128 259:2 0 1M 0 part
```
## Windows instances
You can run the **ebsnvme-id** command to map the NVMe device disk number to an EBS volume ID and device name. By default, all EBS NVMe devices are enumerated. You can pass a disk number to enumerate information for a specific device. The `ebsnvme-id` tool is included in the latest AWS provided Windows Server AMIs located in `C:\ProgramData\Amazon\Tools`.
Starting with AWS NVMe driver package `1.5.0,` the latest version of the `ebsnvme-id` tool is installed by the driver package. The latest version is only available in the driver package. The standalone download link for the `ebsnvme-id` tool will no longer receive updates. The last version available through the standalone link is `1.1.0`, which can be downloaded using the link [ebsnvme-id.zip](https://s3.amazonaws.com/ec2-windows-drivers-downloads/EBSNVMeID/Latest/ebsnvme-id.zip) and extracting the contents to your Amazon EC2 instance to get access to `ebsnvme-id.exe`.
```
PS C:\ProgramData\Amazon\Tools> ebsnvme-id.exe
Disk Number: 0
Volume ID: vol-0d6d7ee9f6e471a7f
Device Name: sda1
Disk Number: 1
Volume ID: vol-03a26248ff39b57cf
Device Name: xvdd
Disk Number: 2
Volume ID: vol-038bd1c629aa125e6
Device Name: xvde
Disk Number: 3
Volume ID: vol-034f9d29ec0b64c89
Device Name: xvdb
Disk Number: 4
Volume ID: vol-03e2dbe464b66f0a1
Device Name: xvdc
```
```
PS C:\ProgramData\Amazon\Tools> ebsnvme-id.exe 4
Disk Number: 4
Volume ID: vol-03e2dbe464b66f0a1
Device Name: xvdc
```
# NVMe I/O operation timeout for Amazon EBS volumes
Most operating systems specify a timeout for I/O operations submitted to NVMe devices.
**Linux instances**
On Linux, EBS volumes attached to Nitro-based instances use the default NVMe driver provided by the operating system. Most operating systems specify a timeout for I/O operations submitted to NVMe devices. The default timeout is 30 seconds and can be changed using the `nvme_core.io_timeout` boot parameter. For most Linux kernels earlier than version 4.6, this parameter is `nvme.io_timeout`.
If I/O latency exceeds the value of this timeout parameter, the Linux NVMe driver fails the I/O and returns an error to the filesystem or application. Depending on the I/O operation, your filesystem or application can retry the error. In some cases, your filesystem might be remounted as read-only.
For an experience similar to EBS volumes attached to Xen instances, we recommend setting `nvme_core.io_timeout` to the highest value possible. For current kernels, the maximum is 4294967295, while for earlier kernels the maximum is 255. Depending on the version of Linux, the timeout might already be set to the supported maximum value. For example, the timeout is set to 4294967295 by default for Amazon Linux AMI 2017.09.01 and later.
You can verify the maximum value for your Linux distribution by writing a value higher than the suggested maximum to `/sys/module/nvme_core/parameters/io_timeout` and checking for the Numerical result out of range error when attempting to save the file.
**Windows instances**
On Windows, the default timeout is 60 seconds and the maximum is 255 seconds. You can modify the `TimeoutValue` disk class registry setting using the procedure described in [Registry Entries for SCSI Miniport Drivers](https://learn.microsoft.com/en-us/previous-versions/windows/drivers/storage/registry-entries-for-scsi-miniport-drivers).
# NVMe Abort command for Amazon EBS volumes
The `Abort` command is an NVMe admin command that is issued to end a specific command that was previously submitted to the controller. This command is typically issued by the device driver to storage devices that have exceeded the I/O operation timeout threshold.
Amazon EC2 instance types that support the `Abort` command by default will end a specific command that was previously submitted to the controller when an `Abort` command is issued to attached Amazon EBS volumes. Amazon EC2 instances that do not support the `Abort` command take no action when an `Abort` command is issued to attached Amazon EBS volumes.
The `Abort` command is supported with:
+ Amazon EBS devices with NVMe device version 1.4 or higher.
+ All Amazon EC2 instances, **except** Xen-based instances types and the following Nitro-based instance types:
+ General purpose: A1 \$1 M5 \$1 M5a \$1 M5ad \$1 M5d \$1 M5dn \$1 M5n \$1 M5zn \$1 M6g \$1 M6gd \$1 Mac1 \$1 Mac2 \$1 T3 \$1 T3a \$1 T4g
+ Compute optimized: C5 \$1 c5a \$1 C5ad \$1 C5d \$1 C5n \$1 C6g \$1 C6gd
+ Memory optimized: R5 \$1 R5a \$1 R5ad \$1 R5d \$1 R5dn \$1 R5n \$1 R6g \$1 R6gd \$1 U-12tb1 \$1 U-18tb1 \$1 U-24tb1 \$1 U-3tb1 \$1 U-6tb1 \$1 U-9tb1 \$1 X2gd \$1 X2iezn \$1 Z1d
+ Storage optimized: D3 \$1 D3en \$1 I3en
+ Accelerated computing: DL1 \$1 G4ad \$1 G4dn \$1 G5 \$1 G5g \$1 Inf1 \$1 P3dn \$1 P4d \$1 P4de \$1 VT1
For more information, see section **5.1 Abort command** of the [ NVM Express Base Specification](https://nvmexpress.org/wp-content/uploads/NVM-Express-1_4-2019.06.10-Ratified.pdf).