Data Processing - Buffer (Batch concept in code)

About

Buffers are catalog containers of a (fixed|variable with max) amount of data or operations.

Buffers live in the memory (ie created by code structure) whereas a blocks lives in the disk.

Without buffer, each read or write request is handled directly by the underlying layer (OS, Database, …) . This can make a program much less efficient, since each such request often triggers other operations that may be relatively expensive.. For instance, for an OS layer

  • disk access,
  • network activity,
  • or some other operation

Furthermore, if requested data were sent as single big chunks, only one user could process it at a time, which obviously would make the application very inefficient and not much fun to use. (ie only one user would download data)

Example of buffered data:

List of buffer implementation

Flush Operation

Flushing the buffer means that you write data to an entity (mostly a file).

Documentation / Reference


Powered by ComboStrap