the superblock
When you open a book you do not start reading random pages; first you check the title page and table of contents to learn what kind of book it is and how it is organized. A disk that holds a file system has its own title page: a small, fixed record near the very start of the volume that describes the whole file system. That record is the superblock (Unix's name) or volume control block (the general name).
The superblock holds file-system-wide metadata — facts about the volume as a whole, not about any one file. Typical contents include: the block size (say 4 KB), the total number of blocks and how many are free, the total number of inodes and how many are free, the locations of the inode table and the free-space bitmaps, a magic number that identifies which file-system type this is, and flags such as whether the volume was cleanly unmounted last time. When the operating system mounts the volume, almost the first thing it does is read the superblock into memory, because every later operation needs these numbers to know where things live and how big the pieces are.
Because the superblock is so important, losing it can make an otherwise intact file system unreadable. For that reason real file systems keep backup copies of the superblock scattered across the volume, so a damaged primary copy can be repaired from a spare. Note the superblock is per-volume: a single physical disk split into three partitions has three separate file systems and three separate superblocks.
Running a tool like dumpe2fs on an ext4 volume prints its superblock: Block size: 4096, Inode count: 1310720, Free blocks: 4892013, Filesystem state: clean, plus the byte locations of backup superblocks. The kernel read exactly this record at mount time.
The superblock is the file system's own table of contents — read first, trusted by everything after.
The superblock describes the volume, not your files; per-file metadata lives in the inode (or file control block). Confusing the two is a common beginner slip.