Using the Parallel NFS (pNFS) SCSI Layout with NVMe

The pNFS SCSI layout is a layout type that allows NFS clients to directly perform I/O to block storage devices while bypassing the MDS. It is specified by using concepts from the SCSI protocol family for the data path to the storage devices. This documents explains how to access PCI Express, RDMA or Fibre Channel devices using the NVM Express protocol using the SCSI layout type by leveraging the SCSI Translation Reference (). This document does not amend the pNFS SCSI layout document in any way, instead of explains how to map the SCSI constructs used in the pNFS SCSI layout document to NVMe concepts using the NVMe SCSI translation reference.

The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in .

The following definitions are provided for the purpose of providing an appropriate context for the reader. The "client" is the entity that accesses the NFS server's resources. The client may be an application that contains the logic to access the NFS server directly. The client may also be the traditional operating system client that provides remote file system services for a set of applications. The "server" is the entity responsible for coordinating client access to a set of file systems and is identified by a server owner.

The SCSI layout definition only references few SCSI specific concepts directly. This document uses the NVMe SCSI Translation Reference document () to provide mappings from these SCSI concepts to NVM Express () concepts that SHOULD be used when using the pNFS SCSI layout with NVMe devices. The NVMe SCSI Translation Reference is used to define the NVMe command and concepts that SHOULD be used to implement the pNFS SCSI layout. Implementations MAY or MAY not use an actual SCSI to NVMe translation layer.

The SCSI layout uses the Device Identification VPD page (page code 0x83) from to identify the devices used by a layout. Section 6.1.4 of explains how an an implementation SHOULD construct a valid Device Identification VPD page based on the NVMe Identify data. Only NVMe devices that support either the EUI64 or NGUID value in the Identify Namespace data SHOULD be used as storage devices for the pNFS SCSI layout, as the methods based on the Serial Number for legacy devices might not be suitable for unique addressing needs.

The SCSI layout uses Persistent Reservations to provide client fencing. For this both the MDS and the Clients have to register a key with the storage device, and the MDS has to create a reservation on the storage device. Section 6.7 of contains a full mapping of the required PERSISTENT RESERVE IN and PERSISTENT RESERVE OUT SCSI command to NVMe commands which SHOULD be used when using NVMe devices as storage devices for the pNFS SCSI layout.

The equivalent of the WCE bit in the Caching Mode Page in is the Write Cache Enable field in the NVMe Get Features command, see Section 6.3.3.2 of . If a write cache is enable on a NVMe device used as a storage device for the pNFS SCSI layout, the MDS must ensure to use the NVMe FLUSH command to flush the volatile write cache.

Since no protocol changes are proposed here, no security considerations apply.

The document does not require any actions by IANA.