In 1990, Weimin Du and Ahmed Elmagarmid proposed quasi serializability as a correctness criterion for concurrency control in heterogeneous distributed database systems (HDDBSs). Unlike traditional serializability, which requires transactions to be executed in a strict order, quasi serializability allows local transactions to be executed independently, as long as global transactions are executed sequentially. This approach ensures transaction consistency without violating local autonomy, as local transactions can operate independently. However, global transactions may introduce interference between local transactions at different sites. To prevent this, the authors suggest scheduling transactions properly, such as serially, to avoid undesirable interference. The paper discusses the conditions under which quasi serializable executions maintain transaction consistency. It defines T-consistency as a state where all transactions interfere with each other properly, and shows that quasi serializable executions maintain T-consistency if the distributed interference relation is acyclic. The authors also introduce the concept of value dependency graphs to ensure acyclicity of distributed interference relations. By maintaining these graphs, the system can prevent cycles that could lead to inconsistencies. The paper concludes that quasi serializability is an effective correctness criterion for concurrency control in HDDBSs, as it allows for the maintenance of transaction consistency without violating local autonomy. It also highlights the importance of controlling remote value dependency to ensure the acyclicity of distributed interference relations. The authors note that further research is needed to address the issue of preserving data integrity in HDDBSs using quasi serializable executions.In 1990, Weimin Du and Ahmed Elmagarmid proposed quasi serializability as a correctness criterion for concurrency control in heterogeneous distributed database systems (HDDBSs). Unlike traditional serializability, which requires transactions to be executed in a strict order, quasi serializability allows local transactions to be executed independently, as long as global transactions are executed sequentially. This approach ensures transaction consistency without violating local autonomy, as local transactions can operate independently. However, global transactions may introduce interference between local transactions at different sites. To prevent this, the authors suggest scheduling transactions properly, such as serially, to avoid undesirable interference. The paper discusses the conditions under which quasi serializable executions maintain transaction consistency. It defines T-consistency as a state where all transactions interfere with each other properly, and shows that quasi serializable executions maintain T-consistency if the distributed interference relation is acyclic. The authors also introduce the concept of value dependency graphs to ensure acyclicity of distributed interference relations. By maintaining these graphs, the system can prevent cycles that could lead to inconsistencies. The paper concludes that quasi serializability is an effective correctness criterion for concurrency control in HDDBSs, as it allows for the maintenance of transaction consistency without violating local autonomy. It also highlights the importance of controlling remote value dependency to ensure the acyclicity of distributed interference relations. The authors note that further research is needed to address the issue of preserving data integrity in HDDBSs using quasi serializable executions.