CalDigit RAIDShield - Manual (Page 33)

32
32
Glossary
• BIOS
An acronym for Basic Input/Output System. It is the boot firmware program on a PC, and controls the computer or the hardware device
(like our CalDigit RAID controller) from the time you start it up until the operating system takes over. When you turn on a PC, the BIOS first
conduct a basic hardware check, and then it loads the operating system into your computer's random access memory, or RAM. If you want to
boot up from CalDigit RAID storage in a PC, you will need to load the BIOS to the controller.
• Boot Code
Provide Input, output, and display signal conversion. It is the code that is first executed when the RAID controller is powered on and
allows firmware to be loaded onto the RAID controller.
• Cache
Controller memory used to speed up data transfer to and from a disk.
• Disk Array
A collection of disks from one or more commonly accessible disk controllers, combined with a body of Array Management Software.
Array Management Software controls the disks and presents them to the array operating environment as one or more virtual disks.
• EFI
EFI is a replacement for the original BIOS firmware. Originally developed by Intel it redefines how firmware communicates with the
operating system. It contains such information as: platform-related details, boot and runtime service calls.
• Host Computer
Any computer system to which disks are directly attached and accessible for I/O. Mainframes, and servers, as well as workstations and
personal computers, can all be considered host computers in the context of this manual, as long as they have disks attached to them.
• LUN
A LUN (Logical Unit Number) is a unique identifier that enables slices to have multiple logical units. Each LUN is a unique number that
identifies a specific logical unit, which may be an end user, a file, an application, a drive, or a volume.
• Native Command Queuing (NCQ)
NCQ allows several outstanding commands to be given to the drives at one time. The commands are carried out in sequence instead of the
order they are given. Rather like pressing buttons in a lift, the lift goes to the next floor in the list not the order that the buttons are pressed.
This speeds up the disk access and reduces the load on the drives.
• Parity
Parity information is redundancy information calculated from actual data values. If any single piece of data is lost, the remaining data and
the parity information can be used together to calculate the lost data. Parity information can either be stored on a separate, dedicated drive, or
be mixed with the data across all the drives in the array.
• RAID (Redundant Array of Independent / Inexpensive Disks)
A disk array in which part of the storage capacity is used to store redundant information about user data stored on the remainder of the
storage capacity. The redundant information enables regeneration of user data in the event that one of the array member disks or the access
path to it fails. See Parity. Different RAID levels offer different data output speeds and fault tolerance (data redundancy). RAID 0 does not
feature redundant information but is nonetheless considered a type of RAID.
• SMART (Self-Monitoring, Analysis and Reporting Technology) Mode
SMART Mode monitors the performance of the hard drives to predict hard drive failure.
• Stripe Size
Stripe size is the maximum number of sectors the RAID system can access without accessing another disk. The stripe size is also the size
of the cache for the RAID. A larger stripe is preferable since it reduces the number of I/O requests made to a physical disk and lets the buffer
cache work more efficiently.
• Slicing
Unlike striping, slicing allows the creation of arrays from a single disk without a loss of speed as the disk fills up. This is because when
striping across disks the center of the disk fills up and when it’s being written to it slows down. Slicing creates new disk partitions with similar
characteristics, therefore keeping the speed the same.