A Non-Technical Look inside the EXT2 File System

How the EXT2 file system is organized on the disk and how it gets its speed.
Group Descriptors and Bitmaps

The next block of each group is the group descriptor. The group descriptor stores information on each group. Within each group descriptor is a pointer to the table of inodes (more on inodes in a moment) and allocation bitmaps for inodes and data blocks.

An allocation bitmap is simply a list of bits describing which blocks or inodes are in use. For example, data block number 123 is in use if bit number 123 in the data bitmap is set. Using the data and inode bitmaps, the file system can determine which blocks and inodes are in current use and which are available for future use.

Inodes and Such

Each file on disk is associated with exactly one inode. The inode stores important information about the file including the create and modify times, the permissions on the file and the owner of the file. The inode also contains the type of the file (regular file, directory, device file like /dev/ttyS1, etc.) and the location of the file on disk.

The data in the file is not stored in the inode itself. Instead, the inode points to the location of the data on disk. There are fifteen pointers to data blocks within each inode. However, this does not mean that a file can only be fifteen blocks long. Instead, a file can be millions of blocks long, thanks to the indirect way that data pointers point to data.

The first thirteen pointers point directly to blocks containing file data. If the file is thirteen or fewer blocks long, then the file's data is pointed to directly by pointers within each inode and can be accessed quickly. The fourteenth pointer is called the indirect pointer and points to a block of pointers, each one of which points to data on the disk. The fifteenth pointer is called the doubly indirect pointer and points at a block containing many pointers to blocks each of which points at data on the disk. The picture shown in Figure 1 should make things clear.

Figure 1. The pointers between an inode and its associated data.

This scheme allows direct access to all the data of small files (files less than fourteen blocks long) and still allows for very large files with only a few extra accesses. As Table 1 shows, almost all files are actually quite small; therefore, almost all files can be accessed quickly using this scheme.

Table 1. Occurrence of Various File Sizes

Inodes are stored in the inode table, which is at a location pointed to by the group descriptor within each group. The location and size of the inode table is set at format time and cannot be changed without reformatting. This means that the maximum number of files in the file system is also fixed at format time. However, each time you format the file system you can set the maximum number of inodes with the -i option to mke2fs.

Directories

No one would like a file system where files were accessed by inode number. Instead, people want to give textual names to files. Directories associate these textual names with the inode numbers used internally by the file system. Most people don't realize that directories are just files where the data is in a special directory format. In fact, on some older Unix systems, you could run editors on the directories, just to see what they looked like internally (imagine running vi /tmp).

Each directory is a list of directory entries. Each directory entry associates one file name with one inode number and consists of the inode number, the length of the file name and the actual text of the file name.

The root directory is always stored in inode number two, so that the file system code can find it at mount time. Subdirectories are implemented by storing the name of the subdirectory in the name field and the inode number of the subdirectory in the inode field. Hard links are implemented by storing the same inode number with more than one file name. Accessing the file by either name results in the same inode number, and therefore, the same data.

The special directories “.” and “..” are implemented by storing the names “.” and “..” in the directory and the inode number of the current and parent directories in the inode field. The only special treatment these two entries receive is that they are automatically created when any new directory is made, and that they cannot be deleted.

Figure 2. Layout of a disk with one partition and four groups.

______________________

Webinar
One Click, Universal Protection: Implementing Centralized Security Policies on Linux Systems

As Linux continues to play an ever increasing role in corporate data centers and institutions, ensuring the integrity and protection of these systems must be a priority. With 60% of the world's websites and an increasing share of organization's mission-critical workloads running on Linux, failing to stop malware and other advanced threats on Linux can increasingly impact an organization's reputation and bottom line.

Learn More

Sponsored by Bit9

Webinar
Linux Backup and Recovery Webinar

Most companies incorporate backup procedures for critical data, which can be restored quickly if a loss occurs. However, fewer companies are prepared for catastrophic system failures, in which they lose all data, the entire operating system, applications, settings, patches and more, reducing their system(s) to “bare metal.” After all, before data can be restored to a system, there must be a system to restore it to.

In this one hour webinar, learn how to enhance your existing backup strategies for better disaster recovery preparedness using Storix System Backup Administrator (SBAdmin), a highly flexible bare-metal recovery solution for UNIX and Linux systems.

Learn More

Sponsored by Storix