Given a pair of locked transactions, accessing a distributed database, the problem is studied of whether this pair is safe, i.e., guaranteed to produce only serializable schedules. It is shown that an easy-to-test graph condition, which characterizes safety for a pair of locked transactions in a centralized database, also applies when the database has been distributed among at most three sites
AbstractIn a multidatabase system, global transactions are executed under the control of the global ...
AbstractIn this paper, we extend the traditional relationships between locks, i.e., shared and non-s...
An important characteristic of concurrency control mechanisms is the level of concurrency that they ...
Given a pair of locked transactions, accessing a distributed database, the problem is studied of whe...
AbstractThe problem of determining whether a set of locked transactions, accessing a distributed dat...
We examine the problem of determining whether a set of locked transactions, accessing a distributed ...
AbstractWe analyze the problem of determining freedom from deadlock of transactions which control co...
AbstractThe purpose of a database concurrency control is to allow only serializable executions of tr...
AbstractIn this paper we compare the performances of concurrency control algorithms using the combin...
Various techniques have been proposed to ensure the safe, concurrent execution of a set of database ...
AbstractIn this paper we study the serializability and the concurrent access to a database by making...
A method for concurrency control in distributed database management systems that increases the level...
Although the general concepts provided by the standard concurrency control theory (e. g. [BHG87]) re...
AbstractDatabases and other transaction-processing systems use concurrency control and recovery algo...
AbstractYannakakis showed that a locking policy is not safe if and only if it allows a canonical non...
AbstractIn a multidatabase system, global transactions are executed under the control of the global ...
AbstractIn this paper, we extend the traditional relationships between locks, i.e., shared and non-s...
An important characteristic of concurrency control mechanisms is the level of concurrency that they ...
Given a pair of locked transactions, accessing a distributed database, the problem is studied of whe...
AbstractThe problem of determining whether a set of locked transactions, accessing a distributed dat...
We examine the problem of determining whether a set of locked transactions, accessing a distributed ...
AbstractWe analyze the problem of determining freedom from deadlock of transactions which control co...
AbstractThe purpose of a database concurrency control is to allow only serializable executions of tr...
AbstractIn this paper we compare the performances of concurrency control algorithms using the combin...
Various techniques have been proposed to ensure the safe, concurrent execution of a set of database ...
AbstractIn this paper we study the serializability and the concurrent access to a database by making...
A method for concurrency control in distributed database management systems that increases the level...
Although the general concepts provided by the standard concurrency control theory (e. g. [BHG87]) re...
AbstractDatabases and other transaction-processing systems use concurrency control and recovery algo...
AbstractYannakakis showed that a locking policy is not safe if and only if it allows a canonical non...
AbstractIn a multidatabase system, global transactions are executed under the control of the global ...
AbstractIn this paper, we extend the traditional relationships between locks, i.e., shared and non-s...
An important characteristic of concurrency control mechanisms is the level of concurrency that they ...