An efficient real-time scheduler for nested transaction models

Nested transaction models have been widely adopted in many advanced database applications, such as telecommunications and real-time traffic information systems. However, noticeable studies focus on scheduling flat transactions for the current research of distributed real-time database systems (RTDBS). This paper presents an efficient real-time scheduler consisting of (1) a priority assignment policy called FHRN to schedule real-time nested transactions, and (2) a lock mechanism called 2PL-HPN to resolve the data conflict problem among nested transactions for distributed RTDBS using nested transaction models. The simulation result shows that the FHRN outperforms current priority assignment policies such as ED, HV and HRU.