Witold Andrzejewski

GPU-WAH: applying GPUs to compressing bitmap indexes with word aligned hybrid

Bitmap indexes are one of the basic data structures applied to query optimization in data warehouses. The size of a bitmap index strongly depends on the domain of an indexed attribute, and for wide domains it is too large to be efficiently processed. For this reason, various techniques of compressing bitmap indexes have been proposed. Typically, compressed indexes have to be decompressed before being used by a query optimizer that incurs a CPU overhead and deteriorates the performance of a system. For this reason, we propose to use additional processing power of the GPUs of modern graphics cards for compressing and decompressing bitmap indexes. In this paper we present a modification of the well known WAH compression technique so that it can be executed and parallelized on modern GPUs.

Graphics processing units in acceleration of bandwidth selection for kernel density estimation

Abstract The Probability Density Function (PDF) is a key concept in statistics. Constructing the most adequate PDF from the observed data is still an important and interesting scientific problem, especially for large datasets. PDFs are often estimated using nonparametric data-driven methods. One of the most popular nonparametric method is the Kernel Density Estimator (KDE). However, a very serious drawback of using KDEs is the large number of calculations required to compute them, especially to find the optimal bandwidth parameter. In this paper we investigate the possibility of utilizing Graphics Processing Units (GPUs) to accelerate the finding of the bandwidth. The contribution of this paper is threefold: (a) we propose algorithmic optimization to one of bandwidth finding algorithms, (b) we propose efficient GPU versions of three bandwidth finding algorithms and (c) we experimentally compare three of our GPU implementations with the ones which utilize only CPUs. Our experiments show orders of magnitude improvements over CPU implementations of classical algorithms.

GPU-Accelerated Collocation Pattern Discovery

Collocation Pattern Discovery is a very interesting field of data mining in spatial databases. It consists in searching for types of spatial objects that are frequently located together in a spatial neighborhood. Application domains of such patterns include, but are not limited to, biology, geography, marketing and meteorology. To cope with processing of these huge volumes of data programmable high-performance graphic cards GPU can be used. GPUs have been proven recently to be extremely efficient in accelerating many existing algorithms. In this paper we present GPU-CM, a GPU-accelerated version of iCPI-tree based algorithm for the collocation discovery problem. To achieve the best performance we introduce specially designed structures and processing methods for the best utilization of the SIMD execution model. In experimental evaluation we compare our GPU implementation with a parallel implementation of iCPI-tree method for CPU. Collected results show order of magnitude speedups over the CPU version of the algorithm.

A Parallel Algorithm for Building iCPI-trees

In spatial databases collocation pattern discovery is one of the most interesting fields of data mining. It consists in searching for types of spatial objects that are frequently located together in a spatial neighborhood. With the advent of data gathering techniques, huge volumes of spatial data are being collected. To cope with processing of such datasets a GPU accelerated version of the collocation pattern mining algorithm has been proposed recently [3]. However, the method assumes that a supporting structure that contains information about neighborhoods (called iCPI-tree) is given in advance. In this paper we present a GPU-based version of iCPI-tree generation algorithm for the collocation pattern discovery problem. In an experimental evaluation we compare our GPU implementation with a parallel implementation of iCPI-tree generation method for CPU. Collected results show that proposed solution is multiple times faster than the CPU version of the algorithm.

Indexing of sequences of sets for efficient exact and similar subsequence matching

Object-relational database management systems allow users to define complex data types, such as objects, collections, and nested tables. Unfortunately, most commercially available database systems do not support either efficient querying or indexing of complex attributes. Different indexing schemes for complex data types have been proposed in the literature so far, most of them being application-oriented proposals. The lack of a single universal indexing technique for attributes containing sets and sequences of values significantly hinders practical usability of these data types in user applications. In this paper we present a novel indexing technique for sequence-valued attributes. Our index permits to index not only sequences of values, but sequences of sets of values as well. Experimental evaluation of the index proves the feasibility and benefit of the index in exact and similar matching of subsequences.

Parallel GPU-based Plane-Sweep Algorithm for Construction of iCPI-Trees

This article tackles the problem of efficient construction of iCPI trees, frequently used in co-location pattern discovery in spatial databases. It discusses the methods for parallelization of iCPI-tree construction and plane-sweep algorithms used in state-of-the-art algorithms for co-location pattern mining. The main contribution of this paper is threefold: 1 a general algorithm for parallel iCPI-tree construction is presented, 2 two variants of parallel plane-sweep algorithm which can be used in conjunction with the aforementioned iCPI-tree construction algorithm are introduced and 3 all three algorithms are implemented on CUDA GPU platform and their performance is tested against an efficient multithreaded parallel implementation of iCPI-tree construction on CPU. Experiments prove that our solutions allow for large speedups over CPU version of the algorithm. This paper is an extension of the conference paper Andrzejewski & Boinski, 2014.

FOCUS: an index for continuous subsequence pattern queries

Recent appearance of the a type of OLAP analysis, the sequential OLAP (or SOLAP) has caused the need for new index structures which support new types of analytical queries. An integral part of processing SOLAP queries is finding sequences which match a user-specified pattern. We call such queries subsequence pattern queries. The contribution of this paper is threefold: first, we propose logical and physical index structure which supports subsequence pattern queries, second, we extend this structure to support aggregation queries and third, we perform performance experiments which show that our solutions offer orders of magnitude improvement over previous state of the art solutions.

AISS: an index for non-timestamped set subsequence queries

In many recent applications of database management systems data may be stored in user defined complex data types (such as sequences). However, efficient querying of such data is not supported by commercially available database management systems and therefore efficient indexing schemes for complex data types need to be developed. In this paper we focus primarily on the indexing of non-timestamped sequences of sets of categorical data, specifically indexing for set subsequence queries. We address both: logical structure and implementation issues of such indexes. Our main contributions are threefold. First, we specify the logical structure of the index and we propose algorithms for set subsequence query execution, which utilize the index structure. Second, we provide the proposition for the implementation of such index, which uses means available in all of the “of the shelf” database management systems. Finally, we experimentally evaluate the performance of the index.

Efficient spatial co-location pattern mining on multiple GPUs

Abstract In this paper, we investigate Co-location Pattern Mining (CPM) from big spatial datasets. CPM consists in searching for types of objects that are frequently located together in a spatial neighborhood. Knowledge about such patterns is very important in fields like biology, environmental sciences, epidemiology etc. However, CPM is computationally challenging, mainly due to the large number of pattern instances hidden in spatial data. In this work, we propose a new solution that can utilize the power of multiple GPUs to increase the performance of CPM. The proposed solution is also capable of coping with the GPU memory limits by dividing the work into multiple packages and compressing internal data structures. Experiments performed on large synthetic and real-world datasets prove that we can achieve an order of magnitude speedups in comparison to the efficient multithreaded CPU implementation. Our solution can greatly improve the performance of data analysis, using widely available and energy efficient graphics cards. As a result, CPM in large datasets is more viable for university researchers as well as smaller companies and organizations.

Two-Echelon System Stochastic Optimization with R and CUDA.

Parallelizing of the supply chain simulator is considered in this paper. The simulator is a key element of the algorithm optimizing inventory levels and order sizes in a two-echelon logistic system. The mode of operation of the logistic system and the optimization problem are defined first. Then, the inventory optimization algorithm is introduced. Parallelization for CUDA platform is presented. Benchmarking of the parallelized code demonstrates high efficiency of the software hybrid.