Crowd simulations are attracting increasing attention from both academia and the industry field and are implemented across a vast range of applications, from scientific demonstrations to video games and films. As such, the demand for greater realism in their aesthetics and the amount of agents involved is always growing. A successful crowd simulation must simulate large numbers of pedestrians' behaviours as realistically as possible in real-time. The thesis looks at two important aspects of crowd simulation and real-time animation. First, this thesis introduces a new data structure called Extended Oriented Bounding Box (EOBB) and related methods for fast collision detection and obstacle avoidance in the simulation of crowd motion in virtual environments. The EOBB is extended to contain a region whose size is defined based on the instantaneous velocity vector, thus allowing a bounding volume representation of both geometry and motion. Such a representation is also found to be highly effective in motion planning using the location of vertices of bounding boxes in the immediate neighbourhood of the current crowd member. Second, we present a detailed analysis of the effectiveness of spatial subdivision data structures, specifically for large-scale crowd simulation. For large-scale crowd simulation, computational time for collision detection is huge, and many studies use spatial partitioning data structure to reduce the computational time, depicting their strengths and weaknesses, but few compare multiple methods in an effort to present the best solution. This thesis attempts to address this by implementing and comparing four popular spatial partitioning data structures with the EOBB.

Parallelization of logic simulation on register-transfer and gate level is a promising way to accelerate extremely time-extensive system simulation processes during the design of whole processor structures. The background of this paper is given by the functional simulator parallelTEXSIM realizing simulation based on the clock-cycle algorithm over loosely-coupled parallel processor systems. In preparation for parallel cycle simulation, partitioning of hardware models is necessary, which essentially determines the efficiency of the following simulation. We introduce a new method of parameterized partition valuation for use within model partitioning algorithms. It is based on a formal definition of parallel cycle simulation involving a model of parallel computation called Communicating Processors. Parameters within the valuation function permit consideration of specific properties related to both the simulation target architecture and the hardware design to be simulated. Our partition valuation method allows performance estimation with respect to corresponding parallel simulation. This has been confirmed by tests concerning several models of real processors as, for instance, the PowerPC 604 with parallel simulation running on an IBM SP2.

No description available.

In an increasing number of use cases, databases face the challenge of managing irregularly structured data. Irregularly structured data is characterized by a quickly evolving variety of entities without a common set of attributes. These entities do not show enough regularity to be captured in a traditional database schema. A common solution is to centralize the diverse entities in a universal table. Usually, this leads to a very sparse table. Although today's techniques allow efficient storage of sparse universal tables, query efficiency is still a problem. Queries that reference only a subset of attributes have to read the whole universal table including many irrelevant entities. One possible solution is to use a partitioning of the table, which allows pruning partitions of irrelevant entities before they are touched. Creating and maintaining such a partitioning manually is very laborious or even infeasible, due to the enormous complexity. Thus an autonomous solution is desirable. In this paper, we define the Online Partitioning Problem for irregularly structured data and present Cinderella. Cinderella is an autonomous online algorithm for horizontal partitioning of irregularly structured entities in universal tables. It is designed to keep its overhead low by incrementally assigning entities to partitions while they are touched anyway during modifications. The achieved partitioning allows queries that retrieve only entities with a subset of attributes easily pruning partitions of irrelevant entities. Cinderella increases the locality of queries and reduces query execution cost.

With rapidly increasing datasets and more dynamic workloads, adaptive partial indexing becomes an important way to keep indexing efficiently. During times of changing workloads, the query performance suffers from inefficient tables scans while the index tuning mechanism adapts the partial index. In this paper we present the Adaptive Index Buffer. The Adaptive Index Buffer reduces the cost of table scans by quickly indexing tuples in memory until the partial index has adapted to the workload again. We explain the basic operating mode of an Index Buffer and discuss how it adapts to changing workload situations. Further, we present three experiments that show the Index Buffer at work.

Since data sizes of analytical applications are continuously growing, many data scientists are switching from customized micro-solutions to scalable alternatives, such as statistical and scientific databases. However, many algorithms in data mining and science are expressed in terms of linear algebra, which is barely supported by major database vendors and big data solutions. On the other side, conventional linear algebra algorithms and legacy matrix representations are often not suitable for very large matrices. We propose a strategy for large matrix processing on modern multicore systems that is based on a novel, adaptive tile matrix representation (AT MATRIX). Our solution utilizes multiple techniques inspired from database technology, such as multidimensional data partitioning, cardinality estimation, indexing, dynamic rewrites, and many more in order to optimize the execution time. Based thereon we present a matrix multiplication operator ATMULT, which outperforms alternative approaches. The aim of our solution is to overcome the burden for data scientists of selecting appropriate algorithms and matrix storage representations. We evaluated AT MATRIX together with ATMULT on several real-world and synthetic random matrices.

No description available.

No description available.

Reporting functions reflect a novel technique to formulate sequence-oriented queries in SQL. They extend the classical way of grouping and applying aggregation functions by additionally providing a column-based ordering, partitioning, and windowing mechanism. The application area of reporting functions ranges from simple ranking queries (TOP(n)-analyses) over cumulative (Year-To-Date-analyses) to sliding window queries. We discuss the problem of deriving reporting function queries from materialized reporting function views, which is one of the most important issues in efficiently processing queries in a data warehouse environment. Two different derivation algorithms, including their relational mappings are introduced and compared in a test scenario.