Databases as a service (DBaaS) transfer the advantages of cloud computing to data management systems, which is important for the big data era. The allocation in a DBaaS system, i.e., the mapping from databases to nodes of the infrastructure, influences performance, utilization, and cost-effectiveness of the system. Modeling databases and the underlying infrastructure as weighted graphs and using graph partitioning and mapping algorithms yields an allocation strategy. However, graph partitioning assumes that individual vertex weights add up (linearly) to partition weights. In reality, performance does usually not scale linearly with the amount of work due to contention on the hardware, on operating system resources, or on DBMS components. To overcome this issue, we propose an allocation strategy based on penalized graph partitioning in this paper. We show how existing algorithms can be modified for graphs with non-linear partition weights, i.e., vertex weights that do not sum up linearly to partition weights. We experimentally evaluate our allocation strategy in a DBaaS system with 1,000 databases on 32 nodes.

The ever-growing demand for more computing power forces hardware vendors to put an increasing number of multiprocessors into a single server system, which usually exhibits a non-uniform memory access (NUMA). In-memory database systems running on NUMA platforms face several issues such as the increased latency and the decreased bandwidth when accessing remote main memory. To cope with these NUMA-related issues, a DBMS has to allow flexible data partitioning and data placement at runtime.

Frequent-itemset mining is an essential part of the association rule mining process, which has many application areas. It is a computation and memory intensive task with many opportunities for optimization. Many efficient sequential and parallel algorithms were proposed in the recent years. Most of the parallel algorithms, however, cannot cope with the huge number of threads that are provided by large multiprocessor or many-core systems. In this paper, we provide a highly parallel version of the well-known Eclat algorithm. It runs on both, multiprocessor systems and many-core coprocessors, and scales well up to a very large number of threads---244 in our experiments. To evaluate mcEclat's performance, we conducted many experiments on realistic datasets. mcEclat achieves high speedups of up to 11.5x and 100x on a 12-core multiprocessor system and a 61-core Xeon Phi many-core coprocessor, respectively. Furthermore, mcEclat is competitive with highly optimized existing frequent-itemset mining implementations taken from the FIMI repository.

Frequent-itemset mining is an important part of data mining. It is a computational and memory intensive task and has a large number of scientific and statistical application areas. In many of them, the datasets can easily grow up to tens or even several hundred gigabytes of data. Hence, efficient algorithms are required to process such amounts of data. In the recent years, there have been proposed many efficient sequential mining algorithms, which however cannot exploit current and future systems providing large degrees of parallelism. Contrary, the number of parallel frequent-itemset mining algorithms is rather small and most of them do not scale well as the number of threads is largely increased. In this paper, we present a highly-scalable mining algorithm that is based on the well-known Apriori algorithm; it is optimized for processing very large datasets on multiprocessor systems. The key idea of pcApriori is to employ a modified producer--consumer processing scheme, which partitions the data during processing and distributes it to the available threads. We conduct many experiments on large datasets. pcApriori scales almost linear on our test system comprising 32 cores.

In this paper, we present a parallel method to evaluate functions on pairs of elements. It is a challenge to partition the Cartesian product of a set with itself in order to parallelize the function evaluation on all pairs. Our solution uses (a) replication of set elements to allow for partitioning and (b) aggregation of the results gathered for different copies of an element. Based on an execution model with nodes that execute tasks on local data without online communication, we present a generic algorithm and show how it can be implemented with MapReduce. Three different distribution schemes that define the partitioning of the Cartesian product are introduced, compared, and evaluated. Any one of the distribution schemes can be used to derive and implement a specific algorithm for parallel pairwise element computation.

In modern IT landscapes, databases are subject to a major role change. Especially in Service-Oriented Architectures, databases are more and more frequently dedicated to a single application. Therefore, it is even more important to reflect the application requirements in their design. Software developers and application experts formulate application requirements in software models. Hence, we obviously need to bridge the gap to the software world and directly derive a database design from the software models used in application development and maintenance. We introduce this concept as model-driven database design. In this paper, we present the architecture principles of a model-driven database design tool and details on the enumeration and evaluation of logical database designs.

The XML support in relational databases and the SQL/XML language are still relatively new as compared to purely relational databases and traditional SQL. Today, most database users have a strong relational and SQL background. SQL/XML enables users to perform queries and updates across XML and relational data, but many struggle with writing SQL/XML statements or XQuery update expressions. One reason is the novelty of SQL/XML and of the XQuery expressions that must be included. Another problem is that the tree structure of the XML data may be unknown or difficult to understand for the user. Evolving XML Schemas as well as hybrid XML/relational schemas make it even harder to write SQL/XML statements. Also, legacy applications use SQL but may require access to XML data without costly code changes. Motivated by these challenges, we developed a method to generate SQL/XML query and update statements automatically. The input is either a GUI or a regular SQL statement that uses logical data item names irrespective of their actual location in relational or XML columns in the database. The output is a SQL/XML statement that queries or updates relational and XML data as needed to carry out the original user statement. This relieves the user and simplifies schema evolution and integration. We have prototyped and tested the proposed method on top of DB2 9.5.

The cardinality estimation in ETL processes is particularly difficult. Aside from the well-known SQL operators, which are also used in ETL processes, there are a variety of operators without exact counterparts in the relational world. In addition to those, we find operators that support very specific data integration aspects. For such operators, there are no well-examined statistic approaches for cardinality estimations. Therefore, we propose a black-box approach and estimate the cardinality using a set of statistic models for each operator. We discuss different model granularities and develop an adaptive cardinality estimation framework for ETL processes. We map the abstract model operators to specific statistic learning approaches (regression, decision trees, support vector machines, etc.) and evaluate our cardinality estimations in an extensive experimental study.