The release of the distributed replicated block device DRBD 9.1.0 has been published, which allows you to implement something like a RAID-1 array formed from several disks of different machines connected over a network (network mirroring). The system is designed as a module for the Linux kernel and is distributed under the GPLv2 license.
The drbd 9.1.0 branch can be used to transparently replace drbd 9.0.x and is fully compatible at the protocol level, configuration files and utilities. The changes boil down to reworking the mechanism for setting locks and are aimed at reducing competition when setting locks in the code responsible for I/O in DRBD. The change made it possible to improve performance in configurations with a large number of CPUs and with NVMe drives, by eliminating a bottleneck that negatively affects performance when a large number of parallel I/O requests are received from different CPU cores. Otherwise, the drbd 9.1.0 branch is similar to the 9.0.28 release.
Recall that DRBD can be used to combine cluster node drives into a single fault-tolerant storage. For applications and the system, such storage looks like a block device that is the same for all systems. When using DRBD, all local disk operations are sent to other nodes and synchronized with the disks of other machines. If one node fails, the storage will automatically continue to operate using the remaining nodes. When the availability of the failed node is restored, its state will be automatically brought up to date.
The cluster that forms the storage may include several dozens of nodes located both in the local network and geographically dispersed in different data processing centers. Synchronization in such branched storages is performed using mesh-network technologies (data spreads along the chain from node to node). Nodes can be replicated both synchronously and asynchronously. For example, locally hosted nodes can use synchronous replication, and for remote hosted sites, asynchronous replication can be used with additional compression and traffic encryption.
Source: opennet.ru