Okay, here are the features of the FSes:
FAT32: File Allocation Table, 32-bit
Introduced: August 1996 (Windows 95 OSR2)
Directory contents: Table
File allocation: Linked list
Bad blocks: Cluster tagging
Maximum volume size: 8 TB
Maximum number of files: 268,435,437
Maximum file size: 4GB
Filenames: 8.3 or 255 characters, ASCII except for "'*/:<>?\|
Attributes: Read-only, hidden, system. volume label, subdir, archive
Permissions: No
Transparent compression: No
Transparent encryption: No
Journaling: No
OS support: just about every OS out there.
NTFS: New Technology File System
Introduced: July 1993 (Windows NT 3.1)
Directory contents: B+ tree
File allocation: Bitmap/Extents
Bad blocks: Bitmap/Extents
Maximum volume size: 256TB currently (16EB theoretically)
Maximum number of files: 4,294,967,295
Maximum file size: 16TB currently (16EB theoretically)
Filenames: 255 characters, ASCII except '\0'
Attributes: Read-only, hidden, system, archive
Permissions: ACL
Transparent compression: LZ77
Transparent encryption: DESX, Triple DES, AES
Journaling: Metadata-only
OS support: Native support in Windows NT 3.1-4.0, Windows 2000 (NT 5.0), Windows XP (NT 5.1 and 5.2), and Windows Vista (NT 6.0.) Partial support in Linux (read support since Linux 2.2.0, rw support in 2.6.14 and newer), read support in FreeBSD, eComStation, MacOS X (10.3 and newer.)
Ext2: Second extended file system
Introduced: January 1993
Directory contents: ?
File allocation: Inodes
Bad blocks: ?
Maximum volume size: 16TB
Maximum number of files: 10^18
Maximum file size: 2TB
Filenames: 255 characters, any byte except \0 and /.
Attributes: ?
Permission: POSIX
Transparent compression: Yes (optional)
Transparent encryption: No
Journaling: No
OS support: Native support under Linux and the BSDs. Read-write support in Windows 2000 and newer with a 3rd-party driver, MacOS X.
Here are the other major FSes and how they stack up:
Ext3: Third extended file system
Introduced: November 2001 (Linux 2.4.15)
Directory contents: Table, tree
File allocation: Bitmap, table
Bad blocks: Table
Maximum volume size: 2TB (1KB block) to 16TB (4KB block)
Maximum number of files: number of bytes in volume/2^13.
Maximum file size: 16GB (1KB block) to 2TB (4KB block)
Filenames: 255 bytes long, all bytes except \0 and /.
Attributes: no access time logging, append-only, synchronous write, no-dump, h-tree, immutable, journal, secure-delete, top, allow-undelete.
Permission: POSIX, ACLs and arbitrary permissions (Linux 2.6 and later.)
Transparent compression: no
Transparent encryption: no, provided by the OS instead.
Journal: metadata and file contents
OS support: Native support in Linux and the BSDs, read-write support in Windows 2000 and newer via a 3rd-party driver.
ReiserFS 3.x
Introduced: 2001 (Linux 2.4.1)
Directory contents: B+ tree
File allocation: Bitmap
Bad blocks: ?
Maximum volume size: 16TB
Maximum number of files: 4,294,967,296
Maximum file size: 8TB
Filenames: 255 bytes, all bytes except \0 and /.
Attributes: modification time, metadata change, access time. Tail packing.
Permissions: POSIX, ACLs, and arbitrary security attributes
Transparent compression: No
Transparent encryption: No
Journal: Metadata or block
OS support: Linux (2.4.1 and newer, 2.6.8 or newer for block journal.) Very limited read-only access via 3rd-party programs in Windows.
XFS
Introduced: 1994 (SGI IRIX 3.5)
Directory contents: B+ trees
File allocation: Extent based
Bad blocks: ?
Maximum volume size: 16TB (32-bit OS,) 9EB (64-bit OS)
Maximum number of files: ?
Maximum file size: 16TB (32-bit OS,) 9EB (64-bit OS)
Filenames: 255 bytes, all bytes except /0
Attributes: UNIX extended attributes
Permissions: POSIX, ACLs, arbitrary
Transparent compression: No
Transparent encryption: No, provided by the OS.
Journaling: Metadata
OS support: SGI IRIX (3.5 and newer), Linux (2.4.25), BSD. No tools to read XFS filesystems in non-UNIX OSes.
HFS+: Hierarchical File System Plus
Introduced: Jan. 19, 1998 (MacOS 8.1)
Directory contents: B* tree
File allocation: B* tree
Bad blocks: B* tree
Maximum volume size: 16EB
Maximum number of files: Unlimited
Maximum file size: 16EB
Filenames: 255 characters, Unicode except :.
Attributes: Color, locked, custom icon, bundle, invisible, alias, system, stationary, inetd, no INIT resources, shared, desktop
Permissions: POSIX, ACLs
Transparent compression: No
Transparent encryption: AES
Journal: Metadata-only
OS support: Native support in MacOS (8.1 and newer), read-write support in Linux.
ZFS: Zettabyte File System
Introduced: Nov. 2005 (Solaris 10)
Directory contents: Extensible hash table
File allocation: ?
Bad blocks: ?
Maximum volume size: 16EB
Maximum number of files: 2^48
Maximum file size: 16EB
Filenames: 255 bytes, any Unicode except /0.
Attributes: POSIX
Permissions: POSIX
Transparent compression: Yes
Transparent encryption: No
Journaling: Block
OS support: Native support in Solaris (10). Very limited experimental support in Linux (userspace filesystem) and BSD.
Other filesystems that you might hear of that are very new are ext4 and Resier4. Reiser4 is not based on ReiserFS 3.x. It has support for transparent compression and encryption as well as an on-line defragmenter and extremely fast handling of very large directory structures. It is not in the stable Linux kernel sources yet and is not supported on any other OS. Its further development is uncertain as the designer, Hans Reiser, has found himself in jail after shooting his estranged wife, and his company (Namesys) is deep in debt. Ext4 is a successor to ext3 and ups the volume size to 1EB and was introduced in the 2.6.19-rc2 Linux kernel.
------------------------------------------------------------------------------------------------------
For your system, I'd recommend using ext2 as it's more scalable and stable than FAT32. NTFS is roughly equivalent to ext3. ReiserFS, ZFS, and XFS are more scalable than NTFS is, as well as have some neat features (tail packing, much less fragmentation) than NTFS does. I personally run XFS on my machine as it's one of the overall fastest Linux FSes out these. [/b]
Facebook
Twitter
Instagram