BizQuery - Data Sheet

Business Challenges

Today, a company’s value centers on its intangible assets. One of the essential components of these assets besides competencies and relationships is information. Information is data organized according to the company’s business strategy and the current state of the market. Hence, properly organized data enable the company to augment its value and to stay competitive in the eyes of potential investors and customers.

Nevertheless, it is not all as far as information is concerned. Dramatic dynamism of the modern market throws a new challenge in addition to the eternal problem of data organization. This challenge focuses on the absolute necessity of keeping business information up to the current state of the market to meet its requirements at any time.

Having accepted these challenges, we face a need for powerful data integration software meeting the following requirements:

  • A model of integrated data should hide all the peculiarities of the underlying data sources in order to provide a unified view of all data, and to prevent the user from knowing technical details that are unnecessary for doing business.
  • The software should allow the user to make changes into the model in a smooth and cheap way to keep pace with the market.
  • The user should have the ability to issue cross-domain queries to get benefit from a great amount of interrelations (correlations) occurring in usual business data, especially if coming from various sources.

ISP/HITECH FZE offers such a software called BizQuery suite, designed to help customers organize their business data and to keep them in touch with the rapidly changing market. Using this product, customers can build robust, scalable and adaptive solutions based on their unique organizational needs.

The BizQuery suite consists of three components: BizQuery Engine (server), front-end runtime modules for providing the end user with GUI interface, and deployment and administration tools.

The BizQuery suite is founded on two major technologies:

  • UML, a standard language used for business modeling. UML has gained wide popularity in consulting firms
  • XML, an implementation technology suitable for presenting almost all kinds of data

Input/Output

The end user interacts with BizQuery via a GUI interface that consists of three parts: forms and graphic map for a declarative searching technique, and catalogs for a navigational one. The GUI interface returns results in HTML that can be rendered using any web browser.

For developers and advanced end users: input – UQL or Quilt queries and output – data in XML format.

For deployment: GUI interface that allows drawing a business model in UML using a user-friendly notation, and to map available data to the model.

For administration: The BizQuery Manager provides a central place for all administration tasks.

Areas of applications

  • XML documents integration
  • XML and relational databases integration
  • Startup solutions
  • Search by metadata

NB:For enterprise scale data integration and searching, the BizQuery suite can be used

as a data integration layer in a portal solution.

Strengths of BizQuery

Business Benefits

  • True integration allows getting a desirable view of business data described in the high level modeling language UML with a end user friendly graphical notation
  • Cost effectiveness of deployment allows to get any desirable view of the business data by building a virtual superstructure over the existing data without modification of their structure (and leaving their structure unchanged). Nevertheless, this is no magic because (1) one should describe his business data in UML in a desirable way (i.e. define a business model – BizModel for short) and (2) map the available data structures on the BizModel. In addition to avoiding modifications of the existing data structure, using these two procedures, you can get the advantages mentioned in the points following below.
  • Flexibility allows easy adaptation to the fast changing business (market) - besides reducing the cost of changes to the lowest point to help keep the business in the picture. This approach makes the business model as dynamic as the business--making alterations into mapping information via a graphical (GUI) interface allows to adapt the business model to the current business environment in fast and extremely flexible way to meet business challenges.
  • Extensibility covers fresh business information easily in the BizModel - BizQuery has been designed to grow with the growth of the business.
    One can attach new data sources and add new entities to BizModel in an ad hoc fashion at any time without disrupting normal work of the system
  • Support of cross-domain queries across multiple data sources (i.e. the user can determine search criteria that cover two or more data sources)
    Support of queries across multiple domains allows getting a horizontal section of business data crossing departments instead of a highly specialized vertical section provided by most of present day systems in order to see enterprise as a whole
  • Support of unified concepts to business modeling and querying data means that user can query data in terms of the BizModel having used for business engineering and system deployment. Thus, one never leaves his usual business environment.

Features

  • Support of open standards ensures ease of programming and effortless integration with the existing applications (XML, XML name spaces, XSLT, XQuiry, OMG XMI, OMG UML, etc.)----Native techniques to manage XML data
  • Accessibility via two level interface offers two ways to query data in terms of XML as well as in terms of the BizModel.
  • Product extensibility-- The properly layered architecture allows replacing the existing modules with the favorite ones and/or adding new ones (i.e. support for third party products). BizQuery comes with all necessary modules and tools. Nevertheless, user has the flexibility to replace them as and when needed.
  • Wild card support allows a simple yet powerful search syntax for the most complicated search criteria
  • XQuery (former Quilt) support appeals to the customer by XQuery’s expressiveness--a fine grain access control
  • Ease of use-- GUI-based administration and deployment tools (step-by-step wizards, drag-n-drop interfaces for mapping)

Automatic but customized generation of three kinds of the end user query interfaces (i.e.

forms, graphic map for querying with the support of a user friendly UML notation, catalogs)

  • Search by metadata – for non-XML sources - description of non-XML data can be presented in XML in correspondence to the BizModel for further searching on these data
  • Easy interaction from any language and environments using SOAP. This open web service interface enables easy integration for your applications written in any language (Java, VB, Perl, C/C++, C#, Python)—a multi-user concurrent access

Limitations

  • Updates and insertions are not supported
  • Restructuring of result data is not supported in BizModel terms but it can be accomplished by Quilt queries and by means of XSLT transformation at the front end side

Comparisons with Competing Products

From a technical point of view, the BizQuery suite is, in essence, an XML-based data integration software. That is why one should compare it with software products providing the same facilities.

Virtual Integration Approach vs. Warehouse Approach

Warehouse technology is usually applied to solve different tasks from those one tries to solve in BizQuery. Though data integration could be done with warehouse approach, it is essentially another kind of integration with different features and possibilities. Typically, warehousing is used in tandem with data mining tools, which are software layers built upon a warehouse. Data mining tools usually analyze interrelations in data that are fordifferent periods of time inferring some useful but not evident facts. Thus if one wants to use data mining tools on a warehouse, one should store historical data (i.e. data for different periods).

So the main idea is to store historical data in warehouse, and then use data mining tools to benefit from analyzing the history of data.

Data from different sources (such as Relational DBMS or XML data repositories orothers) are loaded in a warehouse. During the loading process, data can be transformed into the needed representation (for efficiency). Then warehouse can process queries to the data loaded. Data can be refreshed in the warehouse by means of loading them again from the data sources in some period of time.

The benefits from warehouse integration are:

  • Uniform interface tointegrated (i.e. loaded in warehouse) data is provided. It means - in

essence - that the user can exploit a single query language to query all integrated data

  • Fast access to integrated datais provided due to the fact that data is gathered and stored in one place, and thus queries can be optimized

The main drawbacks of warehouse integration are:

  • The user gets data in results of queries that are out-of-date. This happens because warehouse updating is done in typically long periods of time, due to the fact that loading the data into warehouse is a very resource and time consuming task.
  • It is very hard to change the representation of integrated data. What is actually in the warehouse is a kind of frozen and static snapshot of operational data. Radical change of the data representation means rebuilding the warehouse from scratch.

In contrast to a warehouse, a virtual integration system is a virtual repository for integratingdata without loading them from different origins. Thusdataremain in its native storage while queries are processed on the fly. BizQuery is a virtual integration system.

The benefits from virtual integration are:

  • Uniform interface tointegrated (i.e.stored inoriginal data sources) data is provided. It means - in essence - that the user can work in terms of one query language to query all integrated data.
  • The user always gets uptodate data as a result of queries, assembled on the fly from original data sources
  • The presentation of integrated data can be changed easily,because it doesnot require reloading of data as it is required for warehouse . Flexibility in data restructuring for BizQuery is very high because data transformation and restructuring organically blends into virtual integration paradigm while XML and XML-related technologies provide rich possibilities for achieving this.

The drawbacks from virtual integration are:

  • access to the integrated data is not so fast as when warehouse is used because during query execution data is being built on the fly, also less possibilities on data indexing are available. For this reason the data size limits that pertain to the system is potentially smaller than for warehouses.
  • virtual integration system would hardly be suitable to store historical data but in general it does not mean that virtual integration system can not be used as a basis for analytical applications.

Comparison with leading Products in the Market

One of the most marketable products is eXcelon from Object Design. It allows integrating data by loading them into its internal repository, optimized for storing a large amount of data and querying it in a fast way. Thus, this product is based on the warehouse integration technique and does not offer the advantages of virtual integration (i.e. flexibility and availability of the latest data).

Advantages of the product:

  • High speed of query execution (caching, indexing, load-balancing across multiple servers)
  • Support of data modification (transaction)

Disadvantages:

  • The approach to integration, the product is based on, makes it impossible to deal with relational data
  • Data quickly become outdated

Another product that is closely identifiable with BizQuery is Tamino from Software AG. It supports data storage locally as well as virtual integration of relational data.

Advantages:

  • High speed of query execution (indexing)
  • Support of data modification but in a restricted way
  • Support of relational data integration

Disadvantages:

  • Does not support cross-domain queries
  • Poor data transformation facilities that does not allow the user to present integrated data in any desirable way
  • Does not provide facilities for painless schema changes

Besides, it is worth mentioning that a great amount of XML search engines, such as Frontier from UserLand Software, are also available but do not usually support any integration facility and can only deal with locally stored XML data.

From this brief survey of the most famous XML integration products, one can conclude that they can be mainly thought of as persistent servers, ones that are optimized for information assets with a long or even permanent life span.

The ability of the BizQuery suite to process cross-domain queries to data modeled in UML as well as in XML and change the model in a effortless way, sets it apart from alternative XML management solutions and allows finding answers to the problems that still remain unsolved.

NB: Products such as Tamino or eXcelon can be used for storing XML data locally as a part of the system. See the product extensibility feature above.

The following table summaries comparative characteristics of the three products

Product Features / eXcelon /
Tamino
/
BizQuery
Virtual Integration / N / Y / Y
RDBMS and XML DBMS support / Not virtually / Yes – for RDBMS; Not virtually for XML / Yes virtually
Cross-source Queries Support / N / N / Y
Automatic Application Generation / N / N / Y
Flexibility in Defining Integrating Model / Y / N / Y
Modification of the Model / N / Restricted / Y
Semantic Searching
(support for UML) / N / N / Y
Local Storage / Y / Y / N
Updates / N / N / N
Flat files support / N / N / N

Desktop version for personal use as an organizer for the office products’ data

The modern office products store their data in various formats and usually provide poor facilities for managing them in a unified way. Thus, a vast amount of information having gathered for some period becomes nearly useless because it is too hard to discover what is really needed. It leads to devaluation of information that only seems to be available.

To help to get over these difficulties, the basic principles underlying the BizQuery suite and the experience gained in the field can be employed to implement a system to deal with the office formats. It is quite real because the modern office products tend to use XML as an internal format for the office data. For example, the new version of Microsoft Office called Office XP uses the XML language to present many kinds of data such as Excel and Access files (Word’s doc format is not in XML yet). That is why BizQuery approach can be successfully used to manage the office data to get all its advantages. In this case one can estimate the volume of potential market as hundreds of thousands copies.

Support of workflow

Economic imperatives are forcing enterprises not only to integrate heterogeneous information resources but also to automate mission-critical applications accessing these shared information resources in order to streamline their business processes. This goal can be achieved by applying workflow techniques. One of the essential requirements to such techniques is to provide a high-level specification language to define enterprise activity in business like terms. The BizQuery approach can be naturally extended to meet this requirement.

Up to date the BizQuery suite offers facilities only for vertical data integration and presenting data in business terms. The extended system will allow managing horizontal dynamics to capture business activity. Using UML to define enterprise workflows permits staying in business environment to manage all aspects of the enterprise information. It leads to dramatical increase in the number of possible applications based on the Extended BizQuery suite and the value of the product for the market. It promises to be very appealing to potential customers.

Frequently Asked Questions

  1. What kind of applications your system fits better – OLTP or OLAP?
    Basically OLTP.
  2. Do you support flat files?
    No.
  1. Do you support both DTD and XML schema?
    We support DTD .
  1. What fine grain access control facilities you provide?
    Model grain.
  1. Do you support updating of data?
    No.
  1. What kinds of adapters do you support?
    We support RDBMS adapter (for JDBC compatible RDBMS) and Java XML DBMS adapter.
  1. What extent of flexibility does your system support?
    a) several global models over the same data sources are supported
    b) mapping data source structures into global model can be written as a XQuery expression
    c) visualization of user interfaces can be customized
  1. How mapping between source and target DTDs is accomplished?
    a) expressed in XQuery manually
    b) performed using Visual Schema Mapper
  1. What types of data (type system) do you support?
    Basic types (i.e. Integer, Real, Boolean, String).
  1. What kinds of UML diagrams does BizQuery support?
    Class diagrams that are used to describe structure and semantics of data.
  1. Do you have heavy internal XML storage?
    No.
  1. Is it Web enabled?
    Yes, the system is based on Web technology.
  1. Can constraints on your attributes be joined with ‘or’ and ‘and’ logic?
    Yes.
  2. Do you support data stream processing technique?
    Yes.
  3. Can the query results be represented in XML?
    Yes.
  1. Do you support ASP?
    No, we support JSP

1