It's all in the documentation, really :)
As we mention in the JPA documentation
""The JPA format was conceived as an alternative to ZIP, designed to be extremely suitable for PHP scripts. The trick is that the JPA format doesn't store a checksum for each file - therefore it reduces the processing overhead during archiving - and it doesn't use a "lookup table" (central directory) as ZIP does. Both of these design decisions lead to extremely fast, low resource usage archiving processes.
Tip: It is recommended that you use the JPA format for all of your backups. You can extract JPA files either on your server using Kickstart, or on your desktop using Akeeba eXtract Wizard.""
Conversely, this is what I write of the ZIP archives
The ZIP format is the most well known archive format and is integrated in many operating systems and desktop environments, including Windows™, Mac OS X™, KDE and GNOME.
Warning: The ZIP format requires the calculation of CRC32 checksums for each file added in the archive. This is a resource intensive operation which will slow down your backup and may lead to timeouts when archiving big files on slow hosts. If this happens, your only choice is not to use the ZIP format; use JPA instead. Unfortunately, we can't do anything about it: it is a combined limitation of the ZIP specification, how PHP works and how your server is set up.
Regarding the restoration of the backup archive, be it a ZIP or a JPA, you can always use Kickstart as described in our Quick Start Guide
, the Restoring Backups
section of our User's Guide, as well as our video tutorial