RAID 5 - Striping with Parity

RAID 5 is the most common RAID in use. It requires at least 3 drives. Data blocks are striped across the drives and portion of each drive has parity algorithms of the other drives. Using the parity data, the computer can recalculate the data blocks, should they no longer be available. That means a RAID 5 array can withstand a single drive failure without losing data or access to data. A RAID 5 can be achieved via software or a hardware controller. RAID 5 has some very important advantages. Access to data is very fast and if a drive fails, there is still have access to data, even while the failed drive is being replaced and the data is being rebuilt on the new drive. There are some disadvantages. Write time is a bit slower due to the need to write parity. Also If one of the disks fails and is replaced, restoring the data (the rebuild time) may take hours or even a day or longer. If another disk goes bad during that time, access to data will be lost. RAID 5 systems can also add to a false sense of security. RAID 5 is ideal for critical data that has to be read at a high speed due to the number of users accessing the data source at the same time. If the RAID 5 is diligently maintained, it is awesome for data that absolutely must be available at all times.

RAID 2, RAID 3, RAID 4

All three of these versions of RAID configurations are just the same as a RAID 5 in that they stripe across multiple drives but all of the parity is stored on a separate drive or set of drives.

Data Recovery from a Failed RAID 5

The first thing we do is assess each individual drive. Once the initial problem for each drive has been bypassed, we make a sector by sector clone of each drive using a tool specifically designed for this process. We then take these clones and rebuild the RAID in a virtual environment. This process takes a highly skilled engineer and more specifically designed equipment and software to accomplish. The trick to this is that we often run into situations where a drive has failed and gone unnoticed for hours, days, or even years. The data on that drive is stale and makes the data come up as corrupt. We have to find that drive and eliminate it from the mix. We then have access to data that can be tested for integrity and functionality.