Faq.html

It's a tool set to generate Executable Analytical information stores, that we call Information Hologram.
QueryObject (QO) information store is optimized to enable ultra fast analysis of data from disparate sources:

replaces the data source systems for any kind of analytical task;

occupies a small footprint even in presence of huge data volume;

can be accessed by any kind of BI toolkit;

ensures information integrity and security;

can be detached from the host and become mobile, as it can be stored on laptops and other mobile media, enabling instant access to the desired information by anyone, any time, any place, and without restrictions;

enables interactive and complex analysis;

can be distributed via multiple channels, such as e-mail, Web, LAN, mobile devices, and on physical supports (CD-ROMs, floppies, USB memories, etc.).

What is a QueryObject Information Hologram?

It is a term used to name the analytical informations produced by the system. We call it an information hologram because it has the capability to show any information viewed from any angle, and from any set of dimensions chosen by the user, without any additional processing. (So it is actually more then a classic hologram).

What type of information can be handled and generated by QO?

Information useful for performing Macro through to Micro analysis (Analytical Information) i.e.: Customer Performance Management, Business Performance Management, flow analysis, traffic billing, marketing analysis, customer behaviour analysis, etc.

Why does QueryObject perform and scale so well?

QO technology processes pure information, not data. This is much more efficient.
Using special kinds of dimensions, it allows an even greater improvement in size and performance, when the QO designer knows what he is looking for in the data.
For example, it is easier to store and work on two numbers that represent a last name, and its occurrence count C, then it would be to actually storing the last names C times.
Working with pure information allows our proprietary algorithms to make very efficient use of memory and CPU resources and to solve the sparse matrix issue.

How is it possible to integrate QO in a typical IT architecture?

QueryObject provides “Plug&Play” integration with existing Data Management system and Information Delivery applications.
On the back-end side, QueryObject provides seamless connectivity to any source data (relational, hierarchical, flat, transactional, warehouse, mainframe). On the front-end side, QueryObject provides a deployment interface to leverage any existing or new analysis and reporting tool in the enterprise in a web or client/server infrastructure.
Moreover, QueryObject Engine easily integrates into existing processing infrastructure.

What are the key differentiators between QueryObject and other Information Management Systems (OLAP, ROLAP, Query&Reporting, DW, DBMS)?

Information Management Systems		QueryObject
1.	Requires an engine to process / manage / deliver representation of data. An extreme definition could be that an Info Mgmt is an Interpreter of Information (the information is created at run time, a block of data at a time and is not natively persistent).	1.	Is a precompiled/executable file-set that contains persistent information.
2.	Typically is a relational structure requiring intensive computation to yield results.	2.	Doesn't require an engine to yield results. (Only ODBC driver or C API is needed)
3.	Server bound.	3.	Totally mobile data structure.
4.	Requires links to ancillary data sources.	4.	Doesn't require accessory data (e.g. indexes, views, logical relations, accessory keys, multi-table indexes).
5.	High storage costs.	5.	Enables complex analysis with minimal footprint.
6.	Cannot normally store long term views.	6.	Is able to contain and let the user access multi-year time series.
7.	Often proprietary analysis tool set required.	7	Is open to any standard analysis tool.
8.	Require a lengthy development cycle (some times years).	8.	Rapid time to first results (weeks).

What is the difference between a common multidimensional cube and a QueryObject?

A QueryObject contains all the possible answers to any query in its structures.

It is a standalone information store, that is easy to transport, deploy and query.

It can contain all the critical original information that you would normally have to drill back into the original data source for.

It supports hierarchies attached at deployment time, so changing a hierarchy on an existing cube is very easy.

It gives fantastic performance in the following areas: data to information compression, stable and fast query time, data load time, infostore build time.

The flexible infostore build environment is designed to easily integrate with customer infrastructure.

What is the difference between QO and an ETL system (Ascential, Hummingbird, Informatica, ...)?

QueryObject isn't an ETL system, but complements ETL functionalities: ETL systems load data from an operational to a DB schema. The QueryObject system covers the information supply chain processes (e.g. from the operational to the delivery).

How can QO be considered a product complementing BI toolkits such as Business Objects or Brio?

It can be used as the back-end optimized information store to support ‘speed-of-thought’ analysis on your data using the mentioned tools as a front end.

What front-end tools cab be used to access the QueryObjects? Which have been certified as QO compatible?

Any ODBC compliant front-end tool can be used to access the QueryObjects. We have certified as compatible Brio, Business Objects, Cognos PowerPlay, Crystal Reports, Informatica Power Analyzer, Microsoft Excel, Microstrategy, Pilot, SPSS.

Is it possible to access a QueryObject using a proprietary front-end tool?

Yes, we provide proprietary programmers interfaces in C and Java, as well as standards based interfaces suh as ODBC,JDBC,OLE-DB, XML

How can QO be considered a product complementing MOLAP tools, and OLAP tools in general?

You can use QO for those jobs for which the other tools are not powerful enough, especially in presence of huge amounts of data. Furthermore, you can use QO to facilitate information delivery: think of it as an PDF file for your enterprise data, a live holographic view of different portions of your enterprise data.

Is it possible to use QO without an OLAP system?

Yes, QueryObjects can be built from virtually any relational data source, as well as some non-relational ones.
Moreover, QueryObject system can be used as a universal analytical source for a variety of delivery tools, but can also be used for different kind of applications not requiring analytical tools (e.g. dispute resolution, charging&billing, fraud analysis etc.).


What kind of applications can be developed with QueryObject?
	Business Performances Management, Customer Performance Management, Business Intelligence applications/solutions:
	> To implement, replace or support a Data Mart Infrastructure;
	> To Perform OLAP analysis;
	> To deliver low cost dynamic reports;
	> To publish through any Portal infrastructure (on WEB, CD, DVD, PDA etc.) very large & complex information.
	In Complex Data Mining analysis:
	> To distribute information to other SYSTEMs / DBs / APPLICATIONs infrastructures at low cost and with high performances and low consumption of resources.
	For Data intensive data management computation:
	> CRM; > Bioinformatics; > Flow analysis; > Billing reconciliation.
	To perform multivariate statistics.

How can QO efficiently solve the flow analysis issues?
QueryObject technology is very good at managing up to 15 dimensions, with a product of the number of categories up to 2^64-1. This means you can have two fairly large dimensions that correspond to source and sink points in a flow. Additional dimensions, metrics and keyback columns can also be added.

Which vertical market solutions can be (have been) developed with QO?
a. Telcos; b. Finance (banking, credit cards); c. Stock market; d. Insurances; e. Transportation; f. Utilities; g. Logistics; h. National security; i. Public Administration; j. Healthcare; k. Bioinformatics. l. other data intensive applications.

What are the supported standards?
1. ODBC 2. JDBC 3. OLE-DB 4. XML 5. DES encryption

What are the QO proprietary processes?
Proprietary access API to QO. The entire data load, data processing and infostore creation process is a proprietary combination of using some well known and best-of-breed approaches/algorithms.

The security of the information contained into a QO is preserved also if it is detached from the host server and distributed physically via e-mail or CD-Roms?
Yes, you can use LDAP servers for authenticating user/group logins; if the server is not available, you can fallback on the user accounts stored in the information store’s own security management file.

When you set the KeyBack feature, the QO can be still considered an executable information or does it become a ROLAP tool?
Still executable, the QO hologram contains its own private indexed extract of the data source that is needed to provide the keyback functionality.

How much time does it take to train a user on QueryObject?
Designing and building QueryObjects: 1 week. Using QueryObjects: 1 hour.

Which are the future development plan for QueryObject?
We are further improving our technology in many directions: - XML metadata - New highly compressed ready file format - Even better compression in QueryObjects - Removal of combinatorial limit - Greatly improved performance in hierarchies - Greater data source access capabilities