This Executive Overview consists of selected key paragraphs from throughout this paper.
Data processing requirements will continue to grow for the foreseeable future, yet staff and financial resources for DBMS applications and support will continue to be stretched thin. Managers are awash in data that they need to use in a variety of ways to maintain their competitive advantage. DBMS systems and tools must quickly evolve to manage the forthcoming data explosion.
The obvious solution to the problems facing today's DBMS managers is to use a central "industrial strength" data storage facility such as that provided by a relational DBMS (RDBMS), but one which extends the relational model to incorporate additional data types and operators. This is precisely what THOR has implemented in its Object Relational Database (ORDBMS) model.
THOR's unique ORDBMS design addresses all the issues discussed to this point. But no matter how elegantly designed or how highly optimized for speed, a solution isn't a solution if it doesn't meet other basic criteria. Specifically, DBMS managers and administrators need a solution that is easy to administer, scaleable, extensible, and open.
Only one database system meets all these criteria- THOR.
Because of THOR's unique methods of data distribution and parallel processing, THOR is able to dispense with many time-consuming administrative tasks that other systems require. THOR also implements a variety of features to ensure RAS.
THOR's shared-nothing approach, combined with its unique data-driven implementation, provides processing that is more inherently parallel and easier to distribute among processors than current implementations.
THOR was designed and implemented using object-oriented (OO) design tools and techniques, and is programmed to understand relational operators within this context.
Object capability allows developers not only to create new data types, but also to create new predicates, such as SIMILAR and PARTIAL, for the new data type FINGERPRINT. These predicates are evaluated by invoking calculation functions which the developer supplies, and which reside within the database just as EQUAL functions do today for each of the data types currently supported by SQL.
Another interesting example of how THOR supports new data types is the World Wide Web. Typical web sites offer several types of data, including text, pictures, movies, and sounds. THOR can support this mixed environment, especially when the requirement is for ad-hoc searches and queries.
By providing support for open interfaces, THOR preserves your investment in standard tools and applications. In fact, THOR SQL objectManager supports both standard SQL, through open interfaces such as the ODBC API, and the SYBASE Open Client/Server APIs, including TransactSQL and DB/Net libraries.
The THOR MPP Data Server™ platform consists of four to 288 processing nodes configured in a shared-nothing configuration. THOR was designed from the ground up to be the first MPP solution that is both massively parallel and inherently object oriented. Of all MPP implementations, only THOR's SQL objectManager is truly object-oriented: all data is encapsulated with the behaviors or functions needed to operate on the data.
The software design of THOR is based on a data-driven model of execution, implemented using object oriented technology. Data-driven processing and object orientation combine to yield a product which is truly unique today, but which will be the paradigm for the next generation of implementations by other vendors.
Data-driven processing finds multiple operations that can be undertaken concurrently within the evaluation of one or more expression. Data-driven processing is inherently parallel, and overcomes the limitations of procedural programming, in which each statement must complete execution before the next statement begins. This model lets THOR process and evaluate many queries at the same time, further increasing performance.
THOR's unique method of distributing query initiation tasks among all nodes in the system, coupled with its data-driven and object-oriented approach to query processing at the node level, means that a large number of queries can be processed in parallel (inter-query parallelism). It also means that a single query can be processed in parallel on all nodes at once (intra-query parallelism.) In this way, THOR fully maximizes the parallel capabilities of its hardware and software system.
In summary, THOR was designed to address and resolve some of the most pressing issues facing database managers and administrators today. As you continue your reading and research in this important area of your business, we hope you will compare THOR to every other system you review. We are confident that THOR will come out on top.