The autonomous decentralized DB system (ADDS) has been proposed to avoid thr Single Point Of Failure(SPOF), where the databases are replicated among distributed subsystem nodes, and update operations to the numerical values of the DBs are independently performed at subsystem nodes within their allowances (called Allowance Volumes). If the request is within the AV, it will be satisfied immediately, and the transaction will be finished shortly. If a shortage of the AV in a subsystem node happens, each node interchanges its request among neighbored nodes. If the AV surplus nodes are there in the system, they give their surplus AV to the request node. Thus the total AV are autonomously adjusted in the system, but the requesting node must wait for the AV satisfied. The arrival rates of the update requests to the subsystems are various, so some nodes consume the AV fast, some nodes slowly. Once the AV shortage condition happens in the system, the response time for that node becomes long. No research work for the proactive AV adjustment has been reported, thus the authors propose to proactively and autonomously adjust the AV in the system in order to balance the AV among the nodes and to lessen the number of the AV shortage nodes as possible.