Reasons that can cause a RAID system no longer works are not based on a uniform structure of various RAID levels, but they create, from data and parity data differently depending on the RAID level. During RAID-0 (striping) data on two disks without parity check stores, which are treated as a volume, so RAID-1 mirrors, however, only the data by HDD0, HDD1, again without parity. Technology investor shines more light on the discussion. Traditional RAID systems are designed to performance redundancy, offer but presumed resilience and hochkapazitare availability. The most common used and most widely used at the wide RAID levels are RAID-0, RAID-1 and RAID-5. Of course, there are still many more RAID levels, but these eke out a niche more, although these very well meet other, lesser-known rather different RAID levels, their specific purpose.
The scenarios that may require a RAID data recovery, we respond later. Basic functions of each RAID level: RAID-0, known also as “striping”, summarizes two disks as a medium, E.g. 2 x 750 GB HDD offers but no redundancy and also no parity to 1 x 1. 5 TB,. RAID-1, also called “mirroring”, reflects the data 1:1 hard disk 1 on Disk2, providing Renundanz, but no parity. RAID-5, known as “parity”, also reflects the data and offers consequently redundancy, but with parity. To calculate of the parity is spread across all disks (minimum 3 units) formed a logical group. A data block within the data blocks of a group contains the parity information and the remaining blocks of data within a group this usage data or work data.
Use data in turn are distributed across all disks, the same applies to the parity data. Please note, from RAID-5, at the latest when it comes to the distribution of the parity data, the topic of RAID is somewhat more complicated. Thus, RAID-5 for small and medium-sized enterprises is the No. 1 entry level solution. See is the potential of failure of the RAID, with all data security, but often. Unfortunately too often, administrators rely on the alleged redundancy providing a RAID-5. Because it may happen that not only a hard disk of a RAID-5 can fail, but for example two and so the RAID can be not made back then. Granted, the probability of hierur may be low, but teaches us “Murphy BBs law” that we should expect just the unexpected. Much more frequently encountered reasons for a loss of data on RAID systems were: the server registration is lost accidentally RAID reconfiguration damaged, corrupted, formatted or lost partitions received faulty defective controller of accidental re-init of the RAID array of failed RAID configuration replacement of components within the array, E.g. using wrong disk. Failure due to overvoltage, voltage spikes. Where also always makes a RAID in use, alone already based on the fact that it is a RAID, underlines the importance of the data contained on the RAID. Should now be a RAID and a RAID data recovery are needed, it can be very fast, very expensive.