Analysis and reporting go hand and hand. And at LogiXML we believe there should be analysis alternatives to OLAP. This is important for companies that are limited by resources, i.e., who can’t really afford a full OLAP infrastructure. Users should have a choice in their BI strategies, and we believe in making that choice easy for them, whether it’s for standard reporting, ad hoc reporting, interactive/dynamic analysis, OLAP reporting and so on. Companies shouldn’t be limited to certain types of BI just because it may be cost or resource prohibitive. We believe in providing full-featured, Web-based BI products that run on a unified platform and that are also affordable. And, we have structured our business and product offerings accordingly.

Analysis Capabilities without OLAP

Our unique Analysis Grid allows users to perform analysis very similar to OLAP without the associated complexity and costs. In 2005, we released our first version of this innovative tool for the casual analyst user. We recognized that some companies simply cannot afford a full OLAP infrastructure. In response, we decided to bring this analysis power to any business user easily and cost effectively.

With the Analysis Grid, end users can dynamically manipulate report data directly in their Web browsers. They can create new columns with user-defined formulas, create charts and graphs, create dynamic crosstab tables, create filters and sorts, and more.

This tool addresses the same issues as OLAP but does not require the same level of setup and maintenance and complexity to define OLAP dimensions, measures, cubes and so on. With our tool, an IT person can define an Analysis Grid within minutes, which includes selecting the data source(s), setting up access to specific data columns, enabling interactive features and deploying it to end users over the Web.

How the Analysis Grid Works

A report developer defines an Analysis Grid in our Studio environment (the same tool used to build our standard, managed reports). This takes less than thirty minutes.

The developer simply sets attributes for the Analysis Grid such as security, the data source(s) from which to pull data, and the interactive features to enable for users such as filtering, paging, charting, grouping, adding new calculated columns, aggregating data or all of the above. The developer saves these specifications in a ‘report definition’ XML file, and then deploys it over the Web for immediate use by business users.

That’s it.

The Analysis Grid is very flexible; users can change fields, columns, aggregations, calculations and so on in realtime, on-the-fly. In addition, if changes need to be made to the Analysis Grid, developers just have to edit and re-deploy the report definition; they don’t have to take extensive time to rebuild an entire cube like they would with OLAP.

When a user accesses an Analysis Grid in their browser, data is loaded directly from the database into LogiXML’s powerful Reporting Server to be sliced and diced as needed in the current session. Users can save their current settings (like formulas, groupings, filters, sort order and so on) for subsequent use. They can also add new columns as needed without having to rebuild the report (or rebuild the cube in the case of OLAP). To save a snapshot of current report data, users can export and save their results as needed.

What’s going on behind the scenes?

This on the fly slicing and dicing of data is handled on our Reporting Server with results served instantaneously to end users in their browsers. So, similar to OLAP, you get fast results.

Using in-memory processing for efficient, high performance reporting, the Analysis Grid scales up to provide interactive reporting for any number of users. And, new data is pulled from the database only when needed; users would just run a new instance of the Analysis Grid.

Although the Analysis Grid itself can scale up to support any number of users, an organization’s data servers must be able to respond adequately. We realized that due to these hardware limitations, performance could decrease for users who run multi-million row reports or have many concurrent users. To address this issue, we developed a valuable add-on to the Analysis Grid called Logi TURBO Engine, which is based on InterSystems’ CACHE database. You can read more about how we use this powerful embedded database to support high performance in this article in Forbes magazine.

This processing engine can perform up to ten times faster when reporting demands significantly increase (for example for multi-million-row reports and high numbers of concurrent users). And, it does so cost-effectively.

To offer caching capabilities to increase performance even more (comparable to OLAP), by the end of July 2006, we will also provide caching and pre-calculation capabilities in our new and innovative Logi Mart product. Information will be cached in Logi Mart and will be refreshed either on a pre-defined schedule or upon an event occurring.