This paper surveys performance models for distributed and replicated database systems. Over the last 20 years a variety of such performance models have been developed and they differ in (1) which aspects of a real system are or are not captured in the model (e.g. replication, communication, non-uniform data access, etc.) and (2) how these aspects are modeled. We classify the different alternatives and modeling assumptions, and discuss their interdependencies and expressiveness for the representation of distributed databases. This leads to set of building blocks for analytical performance models. To illustrate the work that is surveyed, we select a combination of these proven modeling concepts and give an example how to compose a balanced analytical model of a replicated database. We use this example to show how to derive meaningful performance values and to discuss the applicability and expressiveness of performance models for distributed and replicated databases. Finally, we compare the analytical results to measurements in a distributed database system.

The way in which data is distributed across the processing elements of a parallel sharednothing architecture can have a significant effect on the performance of a parallel DBMS. Data placement strategies provide a mechanical approach to determining a data distribution which will provide good performance. However, there is considerable variation in the results produced by different strategies and no simple way of determining which strategy will provide the best results for any particular database application. This paper considers five different data placement strategies and illustrates some of the problems associated with the placement of data by studying the sensitivity of the results produced by these different strategies to the changes in a number of environmental factors, such as the number of processing elements participating in database activities and the size of database. The study was conducted by using an analytical performance estimator for parallel database systems, in the co...

this paper we describe the Oracle7 Parallel Database Server cache coherency policy as implemented in the GoldRush MegaServer [7] and develop an analytical cache memory model for it. The model is developed in the context of a parallel database performance evaluation tool, called STEADY. The rest of the paper is organised as follows. Section 2 describes STEADY and shows how the cache memory model relates to other modules of the tool. Section 3 briefly reviews several types of cache coherency policies, and discusses Oracle7 parallel cache management in detail. In section 4 we develop an analytical model of this policy. Finally, in Section 5 we present some preliminary results from our model. II. An Analytical Parallel DBMS Performance Evaluator

. Parallel database systems are generally recognised as one of the most important application areas for commercial parallel systems. However, the task of managing the performance of a parallel database system is exceedingly complex. The initial choice of hardware configuration to support a particular DBMS application and the subsequent task of tuning the DBMS to improve performance rely not only on the way in which the data is structured, but also on how it is fragmented, replicated and distributed across the processing elements of the system. To understand the behaviour of a particular application requires the study of large volumes of performance data. To simplify this process it is essential to provide some means of presenting performance data in a comprehensible form which will aid visualisation. This paper explores some of the issues relating to decision support for the performance management of parallel database systems and describes an analytical capacity planning tool to assist...

An analytical approach to response time estimation in parallel relational database systems has been developed. It is based on a representation of database activity, in which queries are mapped to low-level patterns of resource consumption, capturing the execution logic of relational operators and mechanisms such as pipelined and partitioned execution. Resource usage profiles are mapped to open multi-class queueing networks. Queue waiting times are estimated using a heuristic rule, which labels resources as M/M/1 or M/G/1 queues. From these and the resource usage profile the average response time of a query is obtained. Synchronisation mechanisms such as pipelines between operators and partitioned parallelism are taken into account. The results of the analytical approach are compared against measurements of Informix XPS performance on a parallel system. Index Terms--- performance estimation, analytical model, queueing networks, pipeline parallelism, validation. 1 Introduction Parallel ...

. The vast number of design options in replicated databases requires efficient analytical performance evaluations so that the considerable overhead of simulations or measurements can be focused on a few promising options. A review of existing analytical models in terms of their modeling assumptions, replication schemata considered, and network properties captured, shows that data replication and intersite communication as well as workload patterns should be modeled more accurately. Based on this analysis, we define a new modeling approach named 2RC (2-dimensional replication model with integrated communication). We derive a complete analytical queueing model for 2RC and demonstrate that it is of higher expressiveness than existing models. 2RC also yields a novel bottleneck analysis and permits to evaluate the trade-off between throughput and availability. 1 Introduction Replication management in distributed databases concerns the decision when and where to allocate physica...

Parallel database systems are complex entities. As part of a course in a limited time scale, it is difficult to provide useful practical experience on these systems that provide deep insight into their behaviour and operation. This paper describes a tool for performance prediction which has been developed to aid the visualisation of parallel database systems and which is currently being used to support teaching. It enables students to experiment with different hardware and software configurations and to view the effects of changes on the performance of the system. It provides insight into how data can be placed among the nodes of a parallel machine according to predefined strategies, as well as manually, and provides feedback on the effect of these on throughput and response time. It is able to provide a good appreciation of the concepts in a relatively short period of time.