Areca ARC-1231ML/1261ML/1280ML Manual Do Utilizador

176

An extra physical disk drive in a RAID configuration that a controller 

can use to automatically rebuild a system drive when another drive 

fails. The hot spare drive must have at least as mush capacity as the 

largest disk drive in the array or the rebuild will not start. 

Beyond the different types of RAID, there are two implementation 

forms: hardware-based and software-based. Hardware-based RAID 

is obviously implemented at a physical level, whereas software-based 

RAID is done using the operating system. It is always optimal to have 

RAID done at a physical (hardware) level.

To pull out a component from a system and plug in a new one while 

the power is still on and the unit is still operating.

(Non-Volatile Random Access Memory) A memory unit is equipped with 

a battery so that the data remain even after the main power had been 

switched off. Actually an EEPROM used to store configuration informa-

tion.

A technique used to protect a system from data loss due to faults. 

When parity is implemented (RAID3 and RAID5), data written is logi-

cally XOR’d together to calculate parity, which is stored on the drives 

along with the data. In a system with three drives, the data is written 

to two drives and the calculated parity is stored on a third drive. If one 

drive fails, data on the failed drive is reconstructed from other data 

and parity. 

An advanced version of the PCI bus introduced in 2002. Rather than 

the shared, parallel bus structure of PCI, PCI Express provides a high-