Ipedo: Unifies XML, SQL for XQuery

Earlier this year, Ipedo launched a major enhancement to its XQuery Views capability in Ipedo XIP to provide corporations deeper insight into their corporate data and content held in Oracle, IBM and Microsoft data stores. The Ipedo upgrade is among the first to follow W3C's adoption of XQuery 1.0 as a formal standard. IDN speak with Ipedo CEO Nick Zhang about Ipedo's new capabilities and the promise of XQuery.

Tags: XQuery, Ipedo, Custom, XML, Ipedo XIP, Developer, Data Sources,

Earlier this year, Ipedo launched a major enhancement to its XQuery Views capability in Ipedo XIP to provide corporations deeper insight into their corporate data and content held in Oracle, IBM and Microsoft data stores.

The Ipedo XIP upgrade follows W3C's adoption of XQuery 1.0 as a formal standard, after almost 7 years of deliberations among vendors and user groups. The Ipedo upgrade interoperates with XML Storage in Oracle 10g, DB2 and Microsoft SQL Server. enabling organizations to better access XML data from contract documents, web services, blogs, RSS feeds and more.

Also notable in the Ipedo upgrade are Ipedo's XQuery Views -- a unique data virtualization approach that allows data to be combined without the need to copy, stage and transform it. These XQuery Views also enable queries across combinations of data from remote XML and relational sources.

"Ipedo… provides the flexibility to use XQuery to address XML data, combined with SQL for accessing more conventional relational data, so our clients can get the best of both worlds." Said Ipedo CEO Nick Zhang in a statement. The Ipedo XIP platform combines powerful SQL and XQuery processing, enabling organizations to fuse together information from any database, content source or Web Service, and uses a patented Dual-Core Query Architecture to power a range of business critical applications.

Other new Ipedo XIP features include:
  • XQuery View Caching Allowing XQuery Views results to be cached natively in XML, providing both faster access for users, and post-processing and drilldown using XQuery.
  • Streaming Result Processing Implements query operators that stream data values, ensuring that query evaluation is memory efficient and allowing clients to cursor through results at the lowest computation costs.
  • Publishing of XQuery Views as Web Services - Allows XQuery results to be invoked directly via a web services API, making it easy to create advanced data services. Ipedo XIP can automatically generate the WSDL file needed for a particular XQuery View.
  • Extension Mechanisms for Custom Data Sources - Ipedo's XQuery extension framework enables access to custom data sources, such as in-house applications accessed using custom APIs.

    These enhancements are available immediately for Ipedo XIP version 4.2 on Windows 2000, Windows NT, Sun Solaris, Red Hat Linux and SuSE Linux. Pricing is on a per-CPU basis.

    IDN speak with Ipedeo CEO Nick Zhang about Ipedo's new capabilities and the promise of XQuery.

    Integration Developer News interview with
    Nick Zhang, CEO

    TOPIC: XQuery Solving Business Problems
    IDN: What are some common business problems that IT is solving with XQuery at this early stage?

    Zhang: There are quite a few, actually. Generally, when an organization needs to combine unlike data sources, XML is a great intermediary format, and XQuery is a great way to extract useful information without writing code. So we see combinations of legacy systems with databases, combinations of external data with internal data, and people starting to leverage unconventional sources like RSS feeds for business intelligence.

    TOPIC: Is XQuery Ready Now?
    IDN: Some make the point that unstructured queries will prove interesting sometime in the future, but they are not quite powerful enough today when primarily applied to structured data. What is your view?

    Zhang: Well, we have several large, household names using XQuery today to provide actionable information: BT, HP, and Sun, to name just a few. Clearly, if your data is totally unstructured, then we are talking about search, and we are all familiar with the limitations of search. But with XQuery, we are typically talking about extracting, joining, and transforming XML data that has some level of structure - what some call semi-structured. These can be everything from financial documents like SEC filings, mortgage applications, and derivatives contracts, to newer information flows from Web Services or RSS feeds.

    TOPIC: XQuery Customer Benefits
    IDN: Could you describe for us the customer-centric benefits and features of an Ipedo 'Enhanced XQuery View"? (In other words, what are the types of different "query UI" or "results UI" that users expect with XQuery?)

    Zhang: First, it's important to understand that Ipedo's server provides a data abstraction or virtualization layer via what we call Views. This means customers do not need to copy data or create new data stores just to process data. This, in and of itself, provides cost and time savings.

    Our Views can be accessed using either SQL or XQuery, so a customer can pick the query language that makes most sense. If a developer is building to an XML data model, for, say, some kind of SOA application, then XQuery would probably be a good choice. We provide both a visual drag-and-drop tool that will auto-generate the queries, as well as access to APIs for those who prefer to write the queries.

    One last thing worth mentioning. It's amazing how five or six lines of XQuery can replace tens of lines of code. It's a very powerful and compact language. Customers love it once they start using it.

    TOPIC: Getting Started with XQuery
    IDN: Does Ipedo have recommendations for "starter projects" with XQuery, or places to go where customers can learn more?

    Zhang: There are a few good places to go for information. The W3C XQuery working group has an extensive listing of resources. We at Ipedo have a basic XQuery tutorial in the Developers section of our Web Site, with some sample apps. And there are already half a dozen developer oriented books on XQuery. And the big database companies now all support XQuery, so another good place to look. So plenty of ways to get started.

    TOPIC: 'Customer Data Sources' for XQuery
    IDN What types of data are the most popular 'Custom Data Sources,' and what are their early uses?

    Zhang: These run the gamut, and are typically specific to an industry or a legacy application. A few that we've see recently are molecular structures for a bioinformatics company, calculations from a risk engine at a bank, and workflow data at an insurance company. Most companies have data beyond relational databases and common packaged apps - which we support - so we need to have an extension mechanism.

    Nick Zhang is a successful entrepreneur with over fifteen years of software industry experience. In 2000, he developed the first version of the Ipedo's XML kernel technology, leveraging his extensive software development and data warehousing experience. Before founding Ipedo, Nick served at senior engineering posts at Critical Path, Xeti, VeriFone, and Hewlett-Packard.