Most high-performance software transactional memories (STM) use optimistic invisible reads. Consequently, a transaction might have an inconsistent view of the objects it accesses unless the consistency of the view is validated whenever the view changes. Although all STMs usually detect inconsistencies at commit time, a transaction might never reach this point because an inconsistent view can provoke arbitrary behavior in the application (e.g., enter an infinite loop). In this paper, we formally introduce a lazy snapshot algorithm that verifies at each object access that the view observed by a transaction is consistent. Validating previously accessed objects is not necessary for that, however, it can be used on-demand to prolong the view's validity. We demonstrate both formally and by measurements that the performance of our approach is quite competitive by comparing other STMs with an STM that uses our algorithm.

https://doc.rero.ch/record/18139/files/Riegel_Torvald_-_A_Lazy_Snapshot_Algorithm_with_Eager_Validation_20100430.pdf

A Lazy Snapshot Algorithm with Eager Validation

In this experience report, we present an evaluation of different techniques to manage concurrency in the context of application servers. Traditionally, using entity beans is considered as the only way to synchronize concurrent access to data in Jave EE and using mechanism such as synchronized blocks within EJBs is strongly not recommended. In our evaluation we consider the use of software transactional memory to enable concurrent accesses to shared data across different session beans. We are also comparing our approach with using (1) entity beans and (2) session beans synchronized by a global lock.

Christophe Gête

Papers

TMBean: Optimistic Concurrency in Application Servers Using Transactional Memory