Ali R. Hurson

A taxonomy and current issues in multidatabase systems

A taxonomy of global information-sharing systems is presented, and the way in which multidatabase systems fit into the spectrum of solutions is discussed. The taxonomy is used as a basis for defining multidatabase systems. Issues associated with multidatabase systems are reviewed. Two major design approaches for multidatabases, global schema systems and multidatabase language systems, are described. Existing multidatabase projects and areas for further research are also discussed.<<ETX>>

Automated resolution of semantic heterogeneity in multidatabases

A multidatabase system provides integrated access to heterogeneous, autonomous local databases in a distributed system. An important problem in current multidatabase systems is identification of semantically similar data in different local databases. The Summary Schemas Model (SSM) is proposed as an extension to multidatabase systems to aid in semantic identification. The SSM uses a global data structure to abstract the information available in a multidatabase system. This abstracted form allows users to use their own terms (imprecise queries) when accessing data rather than being forced to use system-specified terms. The system uses the global data structure to match the users terms to the semantically closest available system terms. A simulation of the SSM is presented to compare imprecise-query processing with corresponding query-processing costs in a standard multidatabase system. The costs and benefits of the SSM are discussed, and future research directions are presented.

TF-ICF: A New Term Weighting Scheme for Clustering Dynamic Data Streams

In this paper, we propose a new term weighting scheme called term frequency-inverse corpus frequency (TF-ICF). It does not require term frequency information from other documents within the document collection and thus, it enables us to generate the document vectors of N streaming documents in linear time. In the context of a machine learning application, unsupervised document clustering, we evaluated the effectiveness of the proposed approach in comparison to five widely used term weighting schemes through extensive experimentation. Our results show that TF-ICF can produce document clusters that are of comparable quality as those generated by the widely recognized term weighting schemes and it is significantly faster than those methods

General-purpose systolic arrays

The extension of systolic array architecture from fixed- or special-purpose architectures to general-purpose, SIMD (single-instruction stream, multiple-data stream), MIMD (multiple-instruction stream, multiple-data stream) architectures, and hybrid architectures that combine both commercial and FPGA (field-programmable gate array) technologies is chronicled. The authors present a taxonomy for systolic organizations, discuss each architectures methods of exploiting concurrencies, and compare performance attributes of each. The authors also describe a number of implementation issues that determine a systolic arrays performance efficiency, such as algorithms and mapping, system integration through memory subsystems, cell granularity, and extensibility to a wide variety of topologies.<<ETX>>

Effective clustering of complex objects in object-oriented databases

Clustering is an effective mechanism for retrieving complex objects. Many object-oriented database management systems have suggested variant clustering schemes to improve their performance. Two issues may compromise the effectiveness of a clustered structure , i.e., object updates and multiple relationships. Updates may destroy the initially cIustered structure, and in a multiple relationship environment, clustering objects based on one relationship may sacrifice others. This paper investigates the updating effects and suggests a dynamic reclustering scheme to reorganize related objects on the disk. A cost model is introduced to estimate the benefit and overhead of reclustering. Reorganizations are performed only when the overhead can be justified. For environments in which multiple relationships among objects exist, the paper proposes a leveled clustering scheme to order related objects into a clustering sequence. Our simulation results show that the leveled clustering scheme has a better access time compared with a single-level clustering scheme.

Object-oriented database management systems: evolution and performance issues

Several research prototypes and commercial object-oriented database management systems (DBMSs) that emphasize the representation and manipulation of complex objects are reviewed. It is argued that clustering and buffering schemes tailored to typical complex object operations offer the best near-term means of improving the performance of databases and that research in clustering and buffering should address recent advances in disk technology: optical and parallel disks. The object-oriented DBMSs reviewed are Orion, Iris, GemStone, Encore, Ontos, Versant, and ObjectStore.<<ETX>>

A vertically layered allocation scheme for data flow systems

The asynchronous nature of a data flow model of computation allows the exploitation of maximum inherent parallelism in many application programs. However, one of the major hurdles facing designers of data flow computers is the issue of program allocation. The problem of finding an allocation algorithm which exploits maximum parallelism while minimizing communication overhead belongs to the class of NP-complete problems. This paper proposes a method called the vertically layered allocation scheme which utilizes heuristic rules in finding a compromise between computation and communication costs in a static data flow environment.

Effects of multithreading on cache performance

As the performance gap between processor and memory grows, memory latency becomes a major bottleneck in achieving high processor utilization. Multithreading has emerged as one of the most promising and exciting techniques used to tolerate memory latency by exploiting thread-level parallelism. The question, however, remains as to how effective multithreading is on tolerating memory latency. The performance of multithreading is not only affected by the overlapping of memory latency with useful computation, but also strongly depends on the cache behavior and the overhead of multithreading (e.g., thread management and context-switch costs). In particular, multithreading affects the behavior of caches, and, thus, the overall performance in a nontrivial fashion. To study these issues, this paper presents the Multithreaded Virtual Processor (MVP) model. MVP integrates the multithreaded programming paradigm and a modern superscalar processor with support for fast context switching and thread scheduling. Our studies with MVP show that, in general, the performance improvements are obtained not only by tolerating memory latency but also lower cache miss rates due to exploitation of data locality. However, multithreading creates an additional stress on the memory hierarchy caused by the interference among threads. Also, the dynamic behavior of multithreaded execution hinders the instruction locality that results in a high number of misses in the L1 instruction cache.

Parallel Architectures for Database Systems

Publisher Summary This chapter examines the impact of current technology on the design of special-purpose database machines (DBMs) and provides a survey of the DBMs that focuses on the adaptability of the designs to the current technology. The computer architectures fall into four groups—namely, single instruction stream-single data stream (SISD), single instruction stream-multiple data stream (SIMD), multiple instruction stream-single data stream (MISD), and multiple instruction stream-multiple data stream (MIMD). Database computer (DBC) uses two forms of parallelism: an entire cylinder is processed in parallel and the system performs queries in a pipeline fashion by separate units around two rings. The systems in high very large scale integration (VLSI)-compatible database machines are designed based on the constraints imposed by technology. In general, they are highly parallel with regular and simple architectures. The future database machine designers ought to concentrate on two important issues: (1) the investigation of the effect and benefit of a specialized database operating system on their DBM designs and (2) the optimization of the system throughput rather than improving the response time of a single request.

Data organization and retrieval on parallel air channels: performance and energy issues

Our interest in the global information sharing process is motivated by the advances in communication and computation technologies, the marriage between the two technologies, and the almost limitless amount of information available on the network. Within the scope of the global information sharing process, when a users request (potentially mobile) is directed to public data, broadcasting has been suggested as an effective mechanism to access data. The effectiveness of the schemes to retrieve public data is determined by their ability to reduce the access latency and power consumed by the mobile unit. Various indexing techniques can be used to further improve the effectiveness of retrieving broadcast data.This paper addresses the application of object indexing in parallel broadcast channels. In addition, to further reduce access latency, it proposes several scheduling schemes to order accesses to the data objects on parallel channels. The proposed schemes are simulated, and analyzed. Our simulation results indicate that the employment of indexing scheme and proper scheduling of object retrieval along the parallel channels drastically reduces both the access latency and power consumption at the mobile unit.