Acunu guys explain one piece of their distribution of Apache Cassandra:
our product […] uses a layout known as Randomised Duplicate Allocation (RDA). In the 2-RDA mode, each block of data is duplicated, and the 2 copies allocated at random among the available devices (other schemes can use more than 2 copies, or a variable number of copies depending on the popularity of the data and space constraints).
As opposed to what’s happening in the Hadoop world, Acunu is not just repackaging Cassandra:
The Acunu Core (“Castle”) is at the heart of our distribution for Apache Cassandra. It comprises a rewrite of the Linux storage stack that offloads much of the storage work from Cassandra. […] It includes optimized OS caching and buffering schemes, new storage algorithms with direct access to disks, SSD-aware storage layout algorithms (coming in V2) and an alternative RAID scheme that rebuilds 2TB disks in 30 minutes.
Original title and link: RAID and Acunu Randomised Duplicate Allocation Disk Layout ( ©myNoSQL)