+7 (812) 622 16 80

RAIDIX Data Storage Celebrates RAID’s 30th Anniversary

21.11.2017

RAIDIX celebrates the 30th anniversary of the RAID technology that enables the user to consolidate multiple drives into a logical array for improved redundancy and performance. The concept of RAID (redundant array of independent disks) was first introduced in 1987 by Berkeley University researchers David Patterson, Garth A. Gibson, and Randy Katz.

In June 1988, the scientists presented a paper “A Case for Redundant Arrays of Inexpensive Disks (RAID)” at the SIGMOD conference (“inexpensive” was switched for “independent” in later naming). The primary levels of the RAID specification were RAID 1 (a mirror disk array), RAID 2 (reserved for levels with dedicated Hamming-code parity), RAID 3 and 4 (byte- and block-level striping with dedicated parity), and RAID 5 (block-level striping with distributed parity).

Initially, RAID was deemed a hardware technology, by and large. Indeed, a physical RAID controller is capable of supporting multiple arrays of various levels concurrently. However, a more efficient implementation of RAID is available with the use of software components (drivers). The Linux kernel, for instance, enables flexible management of RAID devices. Leveraging the Linux kernel modules and erasure coding technology, the RAIDIX software developers created a building block for high-performance and fault-tolerant data storage systems based on commodity-off-the-shelf hardware.

RAIDIX operates with the RAID 0, RAID 5, RAID 6, and RAID 10 levels. The patented proprietary RAIDIX algorithms include the unique RAID 7.3 and N+M levels.

RAID 7.3 is a sibling of RAID 6 with double parity, yet the former delivers greater reliability. RAID 7.3 is a level of interleaving blocks with triple parity distribution, allowing the system to restore data in case up to 3 drives fail. RAID 7.3 enables high-performance levels without any additional CPU load, and optimally fits large arrays over 32 TB capacity.

Another patented technology RAID N+M is the level of interleaving blocks with flexible parity that allows the user to choose the number of disks for checksum allocation. RAID N+M requires at least 8 disks and can sustain complete failure of up to 64 drives in the same group (depending on the number of parity disks).

Development of new algorithms and cutting-edge technology inspired by artificial intelligence and machine learning is the core competency of the RAIDIX in-house Research Lab. Over the years, RAIDIX has registered 10+ technology patents in the US. The Lab’s key lines of research involve gradual elimination of obsolete write levels and associated latencies, advancement of predictive analytics, configuration of data storage policies on-the-fly, and more.