A set of userspace tools for managing pools of deduplicated and/or compressed block storage.
VDO (which includes kvdo and vdo) is software that provides inline block-level deduplication, compression, and thin provisioning capabilities for primary storage. VDO installs within the Linux device mapper framework, where it takes ownership of existing physical block devices and remaps these to new, higher-level block devices with data-efficiency capabilities.
Deduplication is a technique for reducing the consumption of storage resources by eliminating multiple copies of duplicate blocks. Compression takes the individual unique blocks and shrinks them with coding algorithms; these reduced blocks are then efficiently packed together into physical blocks. Thin provisioning manages the mapping from LBAs presented by VDO to where the data has actually been stored, and also eliminates any blocks of all zeroes.
With deduplication, instead of writing the same data more than once each duplicate block is detected and recorded as a reference to the original block. VDO maintains a mapping from logical block addresses (used by the storage layer above VDO) to physical block addresses (used by the storage layer under VDO). After deduplication, multiple logical block addresses may be mapped to the same physical block address; these are called shared blocks and are reference-counted by the software.
With VDO's compression, multiple blocks (or shared blocks) are compressed with the fast LZ4 algorithm, and binned together where possible so that multiple compressed blocks fit within a 4 KB block on the underlying storage. Mapping from LBA is to a physical block address and index within it for the desired compressed data. All compressed blocks are individually reference counted for correctness.
Block sharing and block compression are invisible to applications using the storage, which read and write blocks as they would if VDO were not present. When a shared block is overwritten, a new physical block is allocated for storing the new block data to ensure that other logical block addresses that are mapped to the shared physical block are not modified.
This public source release of VDO includes two kernel modules, and a set of userspace tools for managing them. The "kvdo" module implements fine-grained storage virtualization, thin provisioning, block sharing, and compression; the "uds" module provides memory-efficient duplicate identification. The userspace tools include a pair of python scripts, "vdo" for creating and managing VDO volumes, and "vdostats" for extracting statistics from those volumes.
- RHEL9 VDO Documentation
- RHEL8 VDO Documentation
- RHEL7 VDO Integration Guide
- RHEL7 VDO Evaluation Guide
Each branch on this project is intended to work with a specific release of Enterprise Linux (Red Hat Enterprise Linux, CentOS, etc.). We try to maintain compatibility with active Fedora releases, but some modifications may be required and if they become too extensive we may stop providing these updates at any point for a particular branch.
Version | Intended Enterprise Linux Release | Supported With Modifications |
---|---|---|
6.1.x.x | EL7 (3.10.0-*.el7) | |
6.2.x.x | EL8 (4.18.0-*.el8) | Supported Fedora Releases and Rawhide |
8.x.x.x | EL9 (5.14.0-*.el9) | Supported Fedora Releases and Rawhide |
- Pre-built versions with the required modifications for the referenced Fedora
releases can be found
here and can be
used by running
dnf copr enable rhawalsh/dm-vdo
.
The unstable branch provides a snapshot of our most up to date internal code that is targeting the mainline kernel and is currently under test. There are no guarantees of stability from this branch.
VDO was originally developed by Permabit Technology Corp. as a proprietary set of kernel modules and userspace tools. This software and technology has been acquired by Red Hat, has been relicensed under the GPL (v2 or later), and this repository begins the process of preparing for integration with the upstream kernel.
While this software has been relicensed there are a number of issues that must still be addressed to be ready for inclusion into the upstream kernel. To see the progress of this effort, visit the Unstable branch of this project.
In order to build the user-level programs, invoke the following command from the top directory of this tree:
make
After building the user-level programs, they may be installed in the standard locations by invoking the following command from the top directory of this tree, as the root user:
make install
Community feedback, participation and patches are welcome to the [email protected] mailing list -- subscribe here.
This project is currently a stepping stone towards integration with the Linux kernel. As such, contributions are welcome via a process similar to that for Linux kernel development. Patches should be submitted to the [email protected] mailing list, where they will be considered for inclusion. This project does not accept pull requests.
GPL v2.0 or later. All contributions retain ownership by their original author, but must also be licensed under the GPL 2.0 or later to be merged.