RECOVERY SYSTEM FOR MAIN-MEMORY DATABASE
Anurag Gupta, Han-Yin Chen · 1999
In a Main Memory Database (MMDB) the primary copy of the database resides in volatile main memory. This makes MMDB systems more vulnerable to failures compared to the traditional Disk Resident Databases (DRDB). A backup copy of the database is maintained in secondary storage for recovery purposes. Recovery activities like logging, checkpointing and reloading are used to restore the database to a consistent state after a system crash has occurred. In this paper, we implement a recovery scheme with action consistent checkpointing for a main-memory database system and measure the performance of the recovery mechanism as a function of the number of tuples and the number of fields in a tuple. We find that the recovery time increases almost linearly with the increase in the size of the database and that the number of tuples in the database affects the recovery time more than the size of an individual tuple. 1.