This talk will give insights into the challenges of a "shared nothing" clustering storage replication solution has deal with. Besides
the challenges some of the methods and algorithms used in DRBD-8 are
presented: data generation UUDIs, resync-bitmap, the activity log,
detection of conflicting writes and barrier semantics mapping. The
features that became available with the release of DRBD-8.3 will of
course be presented as well: device stacking, checksum based resync.
Which can be used to set up disaster recovery nodes in addition to a
conventional cluster. In the second part of the talk the current state
of affairs on the future evolution (DRBD-9) is given: transfer log on
stable storage, more than two nodes, new ways of identifying the most up
to date data (dagtag). At the end of the talk a few of the benefits of
the new building blocks will get pointed out: Snapshots after the fact
and the "follow the sun" model.