$ Linux Command

Files

split

Split a file into smaller pieces.

filesplitchunkslarge-filesarchive

Additional Notes

split divides a single file into a number of smaller files by line count or byte size. It is useful for sending large logs by email, uploading in chunks, or feeding huge files into tools that prefer smaller inputs.

The output files are named with a prefix plus a suffix (default xaa, xab, ...). Reassembling is as simple as concatenating them back in order.

Syntax

split [options] [file] [prefix]

Parameters

  • options: Flags that choose the chunk size or naming.
  • file: File to split (use - or omit for standard input).
  • prefix: Output filename prefix (default x).

Common Options

  • -l N, --lines=N: Put N lines in each piece.
  • -b N, --bytes=N: Put N bytes in each piece; suffix K, M, G allowed.
  • -a N: Use N characters for the suffix.
  • -d: Use numeric suffixes (00, 01, ...) instead of letters.
  • --additional-suffix=EXT: Append a file extension such as .part.
  • -e: Do not create empty output files.
  • --filter=CMD: Pipe each piece to a command instead of writing a file.

Examples

split -l 1000 big.log

Split into 1000-line files named xaa, xab, and so on.

split -b 100M large.iso part-

Split a 100 MB-per-file ISO with the part- prefix.

split -d -a 3 -b 50M data.zip backup.zip.

Produce backup.zip.000, backup.zip.001, ... using 3-digit numeric suffixes.

split -l 500 access.log /tmp/chunk-

Write chunks into /tmp with a custom prefix.

cat xaa xab xac > big.log

Reassemble the original file from its pieces in order.

Practical Notes

  • Reassembly order matters; keep the default suffix order intact.
  • Use -b with K/M/G for size-based splitting.
  • --filter lets you pipe each piece straight to gzip or sha256sum.
  • Splitting and re-catting is a simple way to work around file-size limits.