Powering Business Intelligence systems
(BI, Analytics)

Business Intelligence (BI or Analytics system) used to consist of consolidating past data, to then analyze this data in order to help make decisions on future actions.
With the advent of technologies and the development of instantaneity, the uses have multiplied: we must now act on actions with immediate results.

The consequences:

• Traditional BI data is cross-checked in real time with the operational data used by applications, services, etc. Thus, we went from a consolidation for the following week, a few days, then 24 hours, to consolidation needs in real time.
  
  In other words: from a batch Business Intelligence system to a "BI on the fly".
  
  
• A silotage of data integration solutions:
  
  • An ETL for BI
  • A tool for data pipelines
  • An EAI or ESB for traditional data exchange needs
  • A framework for APIs / Microservices
  
  

   

In the physical world: with the electronic price tag (electronic shelf label) in a supermarket, you can easily and quickly adjust prices according to traffic, competition, ...

In the e-commerce world: to recover the prices of its competitors and adjust its pricing accordingly

The emergence of ecosystems like Hadoop has made it possible to solve in a different way the constraints related to volumetry or complexity, in particular, the processing of unstructured data. Constituting a reliable storage system, simple and deemed cheap, it is now possible to set up Datalakes to absorb the endless deluge of data (from IOT / connected objects, social networks, ...).

Data mining has been replaced by Big Data technologies that facilitate the use of Machine Learning (ML) as well as Deep Learning.

The challenge for CIOs is to juggle all these technologies and build a coherent and reliable system that allows, as needed, to move from a descriptive BI to a predictive or even prescriptive BI.

Example: in the marketing field, we are increasingly targeting persons, in relation to a given context, using collected data.

Universal mapping

Stambia simplifies the way you design data integration flows. The notion of universal mapping offers a business-oriented vision that does not require technical competence during the mapping phase (conception of relations between objects). The so-called "top-down" approach allows to focus on the "What" and to let the "How to" be automatically generated by the platform.

Unified management of any type of project

This type of mapping makes it possible to respond to any type of data integration need. This offers the ability to handle, with the same skills, Business Intelligence or Big Data projects, as well as real-time oriented projects (streaming, API, Web Services ...). The learning curve of data integration within organizations is facilitated by the use of a universal vision of data transformation.

BI as a service

Stambia's ability to expose complex Web Services in a matter of seconds helps extend the Business Intelligence capabilities of organizations. Indeed, any BI data can be exposed: aggregated data, detail data or repositories or data dimensions.
This feature does not require a complementary tool, thus guaranteeing an exceptional level of responsiveness at almost no cost.

ELT mode as a performance engine

Stambia is a solution that favors the use of transformation delegation, more commonly known as ELT (Extract, Load and Transform).

This tool philosophy provides the best performance.

ELT is also the guarantor of automatic scalability and immediate treatment without additional cost on the side of the data integration solution.

The power of ELT for the Cloud and Big Data

The cloud has become the preferred architecture for IT departments, including platforms for implementing "No-Software" and "No-Hardware" strategies.

The ELT mode provides a lightweight (without a transformation engine) architecture that meets the requirements of cloud initiatives within organizations, ensuring performance and simplicity of architecture.

The logic is the same for Big Data projects. Stambia, through its ELT approach, will take advantage of all the power of Big Data and No SQL platforms without adding complexity to the architecture

Industrialization is a key element of IT projects around data. There are many opportunities for businesses to leverage data with ease and efficiency. On the other hand, once the phases of experimentation and innovation are carried out by the business teams, then comes the industrialization phase.

Stambia is the ideal solution for the industrialization of data flows.

The vision based on templates allows to encapsulate technical know-how and reuse in industrial mode, without the need for users to master the technical details.

Our customers recommend us. For Stambia, this is the greatest reward.

Stambia, before being a technological solution, is a team of experts that are passionate about what they do.

We seek to understand the needs of our users and to offer them the best, thus producing for everyone the expected level of value.

What if you rediscovered with Stambia what it is to have quality software support and a publisher team at your service?

• Cancel & replace mode: In this mode the data is deleted and replenished entirely (useful for initiating data or ensuring consistency of some reference data)
• Incremental Mode: In this mode just the latest changes are taken into account. This type of integration is useful for large data (facts). It can be paired with event detection features on sources to optimize flows and ensure fresh data.
• Historically-powered mode, sometimes called Slow Changing Dimensions (SCD). This mode makes it possible to track the changes in the data over time in order to make further analysis of the evolution of the data.
• Automatic feed of Data Vault models. Many models exist for Business Intelligence systems. Star models or Data Vault modeling are examples. Stambia is compatible with these different models and allows automatic feeding of tables and associated dimensions or satellites.