Creating partitions and file systems in Linux

1. Disk types, units, and partitions

Before partitioning a disk in Linux, it is useful to have some knowledge of hard drive geometry and structure. You also need to know the conventions used by Linux to name drives and partitions.

Understanding the overall partitioning process can help you avoid mistakes and the loss of valuable data. However careful you are, you should still back up all important data before repartitioning a hard drive.

There are three components in a hard disk structure, including

• Disk interface types
• Partitions
• Units

Disk interface types

Hard disks interface with a computer using one of the following main architectures:

• Integrated Drive Electronics (IDE)
• small computer system interface (SCSI)

The IDE interface is the standard used on most modern PCs. As the name suggests, the disk controller electronics are part of the drive hardware, rather than existing separately on the host computer.

Its widespread use is due largely to the fact that it offers reasonable performance at relatively low cost.

SCSI is the second most widely used PC hard disk interface. SCSI drives are most often found in higher-end equipment due to their more advanced data handling, superior performance and reliability, and higher cost.

In Linux, unlike under Windows and some other operating systems, drives attached to the different types of interfaces are differentiated by name. The letters /dev/hd always indicate an IDE hard disk.

For example, /dev/hda indicates the first IDE drive.

The letters /dev/sd are always used for a SCSI hard disk, so /dev/sda indicates the first SCSI hard drive.

Partitions

Partitions are user-defined subdivisions of a physical disk that behave as though they reside on separate physical disks. Their creation improves the efficiency of data transfer and enables multiple operating systems to coexist on a single PC.Partitioning also enables improved organization of a drive’s contents