Generating a single transactional data stream from multiple database logs

A non-transitory computer-readable device having instructions stored thereon that, when executed by at least one primary database server, cause the at least one primary database server to perform operations comprising:

• generating a first system log entry associated with a first unique identifier, the first system log entry corresponding to a first database operation on a first database object of a disk store;
• generating a second system log entry associated with a second unique identifier, the second system log entry corresponding to a committed database operation on a second database object of an in memory row store (IMRS);
• determining a third unique identifier based on the second unique identifier;
• generating an IMRS log entry associated with the third unique identifier, the IMRS log entry corresponding to the committed database operation;
• scanning the first system log entry, the second system log entry, and the IMRS log entry;
• determining a first stream package including the first unique identifier based on the scanning of the first system log entry;
• determining a second stream package including the third unique identifier based on the scanning of the IMRS log entry;
• and
• determining a replay order associated with the first stream package and the second stream package based on the first unique identifier and the second unique identifier.

A replication system, comprising:

• a primary database server with in row memory storage enabled;
• and
• a secondary database server configured to replay log information received from the primary database server;
• wherein the primary database server comprises one or more processors coupled to a memory, the one more processors configured to:
• generate a first system log entry associated a first unique identifier, the first system log entry corresponding to a first database operation on a first database object of a disk store;
• generate a second system log entry associated with a second unique identifier, the second system log entry corresponding to a committed database operation on a second database object of an in memory row store (IMRS);
• determine a third unique identifier based on the second unique identifier;
• generate an IMRS log entry including the third unique identifier, the IMRS log entry corresponding to the committed database operation;
• scan the first system log entry, the second system log entry, and the IMRS log entry;
• determine a first stream package including the first unique identifier based on the scanning of the first system log entry;
• determine a second stream package including the third unique identifier based on the scanning of the IMRS log entry;
• determine a replay order associated with the first stream package and the second stream package based on the first unique identifier and the second unique identifier;
• and
• send the log information to the secondary database server based on the replay order, the log information including the first stream package and the second stream package.