If you use the all asynchronous method, update copy to the remote site is performed using asynchronous copy. With asynchronous copy, because the main site is updated without waiting for updating at the remote site to be completed, there is no impact on the transaction performance at the main site.

However, the possibility exists that the updated content of the files at the main site (update-copy target files) might not be imported to the remote site. Consequently, if a disaster abnormally terminates the HiRDB system at the main site and HiRDB is restarted at the remote site, its state at restart might differ from the state that it was in immediately before the abnormal termination. With the all asynchronous method, therefore, continuity of a service that was running on the main site cannot be guaranteed after restart.

The figure below provides on overview of the all asynchronous method. The table that follows shows the processing method used for copying the update to the remote site (using the all asynchronous method).

Figure 1-3 Overview of the all asynchronous method

Table 1-3 Processing method used for update copying to the remote site (using the all asynchronous method)

Files copied to the remote site		Processing method used for update copying
Database files		Asynchronous copy
System files	System log files
	Synchronization point dump files
	Status files