The fundamental principles employed in fault-tolerant systems are redundancyin system components and modularization of the design.

These two concepts areutilized in typical systems by means of failstop modules (also called fail-fast [Gray,1985]) and process pairs. A fail-stop module constantly monitors itself, and when itdetects a fault, shuts itself down automatically [Schlichting and Schneider, 1983].Process pairs provide fault tolerance by duplicating software modules. The idea isto eliminate single points of failure by implementing each system service as twoprocesses that communicate and cooperate in providing the service. One of theseprocesses is called the primary and the other the backup. Both the primary and thebackup are typically implemented as fail-stop modules that cooperate in providinga service. There are a number of different ways of implementing process pairs,depending on the mode of communication between the primary and the backup. Local Recovery Management The local recovery manager (LRM) maintains the atomicity and durability properties oflocal transactions at each site. Architecture Volatile storage: The main memory of the computer system (RAM) Stable storage A storage that never looses its contents In reality this can only be approximated by a combination of hardware (non-volatilestorage) and software (stable-write, stable-read, clean-up) componentsDDB 2008/09 J. Gamper Page

Two ways for the LRM to deal with update/write operations In-place update Physically changes the value of the data item in the stable database As a result, previous values are lost Mostly used in databases Out-of-place update The new value(s) of updated data item(s) are stored separately from the oldvalue(s)

 Periodically, the updated values have to be integrated into the stable DBDDB 2008/09 J. Gamper Pa