An archive is a system to which inactive data is transferred. Inactive means, in any case, that the data is in its final state and therefore not changed anymore. Often, but not always, it also means that it no longer needs to be accessed frequently. Each record in an archive should have a retention period assigned, which defines exactly how long it needs to be retained. If data has to be stored for a very long period of time, it might purposely be changed upon transmission to the archive – for example by converting it into a format suitable for long term preservation. The archive’s aim is not to restore the data in its original form or system, but to simply make the information available upon request. This fact is also expressed by referring to the process of getting information from an archive as retrieval and not restore. So, if you put a Microsoft Word document into an archive and want to view it ten years later, the archive is likely to provide a TIFF or PDF/A file, which resembles your original document.
An archive is a system that stores finalized records independent from the system they originated from to ensure preservation.