A transaction signaling method in data replication and synchronization that detects the transaction, reads the signal, and based on the signal value it either simply clears the signal or sends a signal to each of its remote databases in the cluster, tells the remote database that next transaction is originated from the signaling layer of another database, and applies the same transaction to the remote database. Unlike current technologies, this technology allows updates at all databases in the cluster simultaneously. It provides users with a method to not only scale out the reads, but also scale out the updates. It synchronizes the databases in the cluster in real-time and eliminates the need of database backups. It not only increases the availability of data but also offers zero data loss when a disaster happens. It may also improves the performance on updates because it eliminates the overhead of cache transfer existed on conventional clustering technologies such as Oracle Real Application Cluster that share the same storage.