Data management, not database, issues now dominate discussions about extracting information from log files, finding ways to manipulate without delay data in financial transactions, and making sense of telemetry data flows.
HiQube is a new high-performance business intelligence (BI) software solution that quickly delivers in-depth business analysis capability and superior reporting, as a result of its unique HiQube technology. HiQube technology is easy to use and is the first to combine hierarchical, relational and multidimensional database technologies. In doing so, it delivers users with unparalleled decision-making power. HiQube BI software solutions are available and supported worldwide.
In January 2007, Altair Engineering (Ann Arbor, Michigan) purchased Hicare, renamed the company, and began marketing the company’s technology more aggressively. Altair is a privately-held firm with an estimated $140 million in revenue in 2007.
The company’s official Web site is here. I found the pre-acquisition Web site more useful. It is here. Don’t let the Italian descriptions throw you. Google Translate makes short work of the language barrier.
What I find interesting is that innovations are coming from specialist firms often based outside the United States. I wrote about the Canadian outfit Infobright last week, now I want to talk about the Italian company HiQube (formerly HiCare).
The company’s technology is a proprietary database that combines three data management technologies in one system. You can manipulate data in a traditional relational form. You also can implement hierarchical data management. The approach I find most interesting is the company’s multi dimensional (n-cube) data managemnet system. If you are not sure of the differences among these types, let me offer a greatly simplified comment about each type:
- Relational–The Codd database. Think of a table in DB2, Oracle, or SQLServer. Data reside in columns and rows.
- Hierarchical–This is a structure in which records are organized as a tree or parent child relationship. Each child type is related to only one parent type
- Multidimensional–This is a data structure that has three or more independent dimensions; for instance, sales by region by product and by time.
HiQube developed what it called the Lilith Enterprise and Web Server business intelligence software, a decision-making support system with unparalleled graphing and reporting capabilities for interactive visualization of information. Lilith, which reminded me of a character on a popular US television show, provides the ability to view and analyze captured data from multiple perspectives and user profiles. The system is available for on premises installation. You can also use the technology as a Web service.
HiQube’s new tag line is “Elevating data to knowledge”. Instead of exporting relational data from a table, transforming it, and reimporting it into a hierarchical form, you can flip from data management approach to data management approach without these steps. The HiQube system implements what the company calls a “business intelligence supply chain”. The idea is that the system imposes a sequence of actions upon the analyst. The pipeline approach minimizes the chance for error when moving through an analytic process. The six stages are importing data, data staging, calculating, simulations, building dashboard, and broadcasting the outputs. These last two steps allow an analyst to create a point-and-click report and assembling multiple reports into a heads-up, graphical display.
The HiQube system is able to ingest files in different formats. HiQube 5, the current version, can accept sparse and dispersed data in different formats. The system’s import tools provide controls to import only the data required for the analysis. Formats supported include
- Text
- Spreadsheet data
- Data from an open system
- SAP R/3 data and Matlab files.
HiQube’s transformation tools are interesting. These can integrate and roll up data during the import process. Once you have your data in its system you can manipulate the data in hierarchies, relationships, and dimensions.
A typical process can manipulate large amounts of data. HiQube’s presentations reference 22 million cells of data. You can view the data in hierarchies; for example, months, cities, and products. You can define formulae on the fly. Calculations are performed in real time. The system permits interactive “what if” analyses.
Other features of the system include:
- Support for multiple inter-connected qubes and virtual qubes
- Automatic qube reshaping and resizing
- Numeric and non-numeric data types
- Unlimited multi-dimensional calculations and write back functions
Analysts can build their own applications on top of a HiQube data set. HiQube provides more than 25 original charts Each is connected to the data engine. Any graphical object is browseable and drillable. You can animate chaarts if additonal sizzle is requried for a senior management presentation. Taking a page from enterprise publishing systems such as InfoPrint, HiQube can publish reports to the Web or generate a file suitable for traditional printing.
You can obtain a price quote from the company, which is based in Ann Arbor. The engineering team is in Torino, Italy. The restaurants alone are worth a tire-kicking trip.
Stephen Arnold, May 25, 2008