On instruction and data prefetch mechanisms

A description is given of the software life cycle support environment (SLCSE), a VAX-based software development environment. The project was 80% Ada, subject to DOD-STD-2167 reviews and documentation. SLCSE adopted an incremental-build/rapid-prototyping methodology similar to Boehm's spiral model. Eight prototype builds (and the final product) were delivered during the course of the development effort. At the heart of the system is an integrated database implemented via a commercial relational database management system overlaid with an entity-relationship layer that models the entire information content of DOD-STD-2167A. The lessons learned from this project are discussed. Among these lessons are the findings that incremental builds help to prove the viability of critical design approaches and that the use of a spiral-type model requires a new mindset. >

On-line analytical processing (OLAP) is a technology that encompasses applications requiring a multidimensional and hierarchical view of data. OLAP applications often require fast response time to complex grouping/aggregation queries on enormous quantities of data. Commercial relational database management systems use mainly multiple one-dimensional indexes to process OLAP queries that restrict multiple dimensions. However, in many cases, multidimensional access methods outperform one-dimensional indexing methods.We present an architecture for multidimensional databases that are clustered with respect to multiple hierarchical dimensions. It is based on the star schema and is called CSB star. Then, we focus on heuristically optimizing OLAP queries over this schema using multidimensional access methods. Users can still formulate their queries over a traditional star scheme, which are then rewritten by the query processor over the CSB star. We exploit the different clustering features of the CSB star to efficiently process a class of typical OLAP queries. We detect special cases where the construction of an evaluation plan can be simplified and we discuss improvements of our technique.

An overview of the Object Protocol Model (OPM) and the OPM data management tools

SVT1Schema Validation Tool for Microsoft SQL-Server

In this paper, we show how we use Nvidia GPUs and host CPU cores for faster query processing in a DB2 database using BLU Acceleration (DB2's column store technology). Moreover, we show the benefits and problems of using hardware accelerators (more specifically GPUs) in a real commercial Relational Database Management System(RDBMS).We investigate the effect of off-loading specific database operations to a GPU, and show how doing so results in a significant performance improvement. We then demonstrate that for some queries, using just CPU to perform the entire operation is more beneficial. While we use some of Nvidia's fast kernels for operations like sort, we have also developed our own high performance kernels for operations such as group by and aggregation. Finally, we show how we use a dynamic design that can make use of optimizer metadata to intelligently choose a GPU kernel to run. For the first time in the literature, we use benchmarks representative of customer environments to gauge the performance of our prototype, the results of which show that we can get a speed increase upwards of 2x, using a realistic set of queries.

The improper use of commercial data base management systems (DBMSs) has been a major problem throughout the history of data base management. With the emergence of microcomputers, the selection of the appropriate DBMS for a user' s particular applications environment can be difficult. This column reviews and discusses approaches toward evaluating a DBMS and provides examples that illustrate the application of these approaches to microcomputer DBMSs. IS managers would benefit by following this column' s three-step evaluation framework

Although central processor speeds continues to improve, improvements in overall system performance are increasingly hampered by memory latency, especially for pointer-intensive applications. To counter this loss of performance, numerous data and instruction prefetch mechanisms have been proposed. Recently, several proposals have posited a memory-side prefetcher; typically, these prefetchers involve a distinct processor that executes a program slice that would effectively prefetch data needed by the primary program. Alternative designs embody large state tables that learn the miss reference behavior of the processor and attempt to prefetch misses.This paper proposes Content-Directed Data Prefetching, a data prefetching architecture that exploits the memory allocation used by operating systems and runtime systems to improve the performance of pointer-intensive applications constructed using modern language systems. This technique is modeled after conservative garbage collection, and prefetches likely virtual addresses observed in memory references. This prefetching mechanism uses the underlying data of the application, and provides an 11.3% speedup using no additional processor state. By adding less than ½% space overhead to the second level cache, performance can be further increased to 12.6% across a range of real world applications.

Although central processor speeds continues to improve, improvements in overall system performance are increasingly hampered by memory latency, especially for pointer-intensive applications. To counter this loss of performance, numerous data and instruction prefetch mechanisms have been proposed. Recently, several proposals have posited a memory-side prefetcher; typically, these prefetchers involve a distinct processor that executes a program slice that would effectively prefetch data needed by the primary program. Alternative designs embody large state tables that learn the miss reference behavior of the processor and attempt to prefetch likely misses.This paper proposes Content-Directed Data Prefetching, a data prefetching architecture that exploits the memory allocation used by operating systems and runtime systems to improve the performance of pointer-intensive applications constructed using modern language systems. This technique is modeled after conservative garbage collection, and prefetches "likely" virtual addresses observed in memory references. This prefetching mechanism uses the underlying data of the application, and provides an 11.3% speedup using no additional processor state. By adding less than ½% space overhead to the second level cache, performance can be further increased to 12.6% across a range of "real world" applications.

There are two main types of security in database systems: discretionary security and mandatory security. Discretionary security restricts access to data items at the discretion of the owner. Most commercial database management systems (DBMS) employ some form of discretionary security by controlling access privileges and modes of data users (Griffiths & Wade, 1976). Discretionary security is not adequate in a multilevel secure environment, however, because it does not prevent Trojan horse attacks and provides a low level of assurance. Mandatory security, on the other hand, restricts access of data items to cleared database users. It is widely exploited in military applications and provides a high assurance.

Commercial database management systems (DBMSs) have historically seen very limited use within the scientific computing community. One reason for this absence is that previous database systems lacked support for the extensible data structures and performance features required within a high-performance computing context. However, database vendors have recently enhanced the functionality of their systems by adding object extensions to the relational engine. In principle, these extensions allow for the representation of a rich collection of scientific datatypes and common statistical operations. Utilizing these new extensions, this paper presents a study of the suitability of incorporating two popular scientific formats, NetCDF and HDF, into an object-relational system. To assess the performance of the database approach, a series of solution variables from a regional weather forecast model are used to build representative small, medium and large databases. Common statistical operations and array element queries are then performed using the object-relational database, and the execution timings are compared against native NetCDF and HDF operations.

The paper describes a belt design package which works from within a commercial Computer Aided Design and Drafting package (AutoCAD) environment and utilizes FORTRAN programs for design and selection of lowest weight components for the drive system. The components used in the process are available as stock items in U.S.A. The relevant information on these products is stored in commercial database management systems such as EXCEL and LOTUS 1-2-3. Output from the package consists of scaled drawing and tabular specifications.

SQL assertions are a powerful means used to specify cross-row constraints, and have been available in the SQL standard since 1992. Unfortunately, assertions are not supported in commercial database management systems. Although triggers and application programs can be efficiently used to constrain database content, they are more complex to write and more error-prone. The objective of this paper is to analyze whether the use of materialized views could be a viable solution as regards the automatic implementation of SQL assertions. Materialized views are views that physically store the result of a query and are periodically updated. The method consists of defining a materialized view which contains the number of tuples that violate the condition expressed in the assertion. The materialized view will contain a CHECK constraint that guarantees that the number of tuples that violate the assertion is equal to zero. The proposed method is an easy and automatic means of implementing the integrity constraints described using assertions. We have carried out a series of tests, and although triggers perform better than materialized views in most situations, there are some in which materialized views would be an efficient option. They are easily automatable and less error prone than triggers.

: The Geospatial Information DataBase (GIDB(Trademark)) offers a fully Object Oriented (OO) database approach to managing the input, storage, retrieval, and presentation of geospatial data in relation to a specific user defined area of interest (AOI). It uses both public domain and commercial OO database management systems (DBMS) technology to store and retrieve the data and can present this information in both 2D and 3D perspective views. It also offers the flexibility to query spatial data with reference to time and space.

This paper addresses the problem of using a commercial database management system (DBMS) in an academic environment for educational purposes. Prior to the decision to use a commercial DBMS, an instructional database IDBS [4] was used. The pros and cons of this decision are discussed. In addition, the problems of obtaining a commercial DBMS on an academic budget for instructional purposes, and of creating a reasonably realistic database are discussed. Finally, an evaluation of the initial quarter's use is presented.The environment for this discussion is the Georgia State University College of Business Administration Information Systems Department graduate database course. This course is offered three times a year with approximately 20 students per offering.DBMS are extensively being used in business areas [2,9,14,15]. As more DBMS are being installed, the sophistication and complexity of these systems is increasing [9,12,13]. This fact makes it imperative for the graduates in the Information Systems field to have a solid understanding of a DBMS and how it is used in a business environment. Without actual experience using a DBMS one cannot fully understand systems of such complexity. This understanding cannot be obtained from text reading and lecture. This situation is analogous to learning a programming language without actually ever using a computer.

Although processing speed, storage capacity and network bandwidth are steadily increasing, network latency remains a bottleneck for scientists accessing large remote data sets. This problem is most acute with n-dimensional data. Grid researchers have only recently begun to develop tools for efficient remote access to n-dimensional data sets. Within the context of the Granite Scientific Database system, we show that latency penalties can be dramatically reduced using explicit knowledge of a user's access pattern represented as an iterator. The iterator not only performs an n-dimensional iteration for the user, but also communicates the access pattern to Granite so that a prefetching cache can be constructed that is tuned to the user's access pattern. We experimentally evaluate a scenario for incorporating Granite's prefetching mechanism into the grid, demonstrating extraordinary performance gains. In light of these results, we describe planned additions to existing grid services to allow selection of datasets according to the user access pattern