Bounded Inconsistency for Type-Specific Concurrency Control
Distributed and Parallel Databases
| Hi-index | 0.00 |
Epsilon serializability (ESR) is a weaker form of correctness designed to provide more concurrency than classic serializability (SR) by allowing, for example, query transactions to view incon- sistent data in a controlled fashion $i.e.$ limiting the incon- sistency within the specified bounds. In the previous literature on ESR, inconsistency bounds have been specified with respect to transactions or with respect to objects. In this paper, we in- troduce the notion of hierarchical inconsistency bounds that al- lows inconsistency to be specified at different granularities. The motivation for this comes from the way data is usually organ- ized, in hierarchical groups, based on some common features and interrelationships. Bounds on transactions are specified at the top of the hierarchy, while bounds on the objects are specified at the bottom and on groups in between. We also discuss mechan- isms needed to control the inconsistency so that it lies within the specified bounds. While executing a transaction, the system checks for possible violation of inconsistency bounds bottom up, starting with the object level and ending with the transaction level. Thus far, to our knowledge, no work has been done to determine the quantitative performance improvement resulting from ESR. Hence in this paper we report on an evaluation of the performance improvement due to ESR incorporating hierarchical inconsistency bounds. The tests were performed on a prototype transaction pro- cessing system that uses timestamp based concurrency control. For simplicity, our implementation uses a two level hierarchy for inconsistency specification - the transaction level and the ob- ject level. We present the results of our performance tests and discuss how the behavior of the system is influenced by the tran- saction and object level inconsistency bounds. We make two im- portant observations from the tests. First, the thrashing point shifts to a higher multiprogramming level when transaction incon- sistency bounds are increased. Further, for a particular mul- tiprogramming level and a particular transaction inconsistency bound, the throughput does not increase with increasing object inconsistency bounds but peaks at some intermediate value. --------------- This material is based upon work supported by the National Science Foundation under grant IRI-9109210. ******************************************************************************