icCube
Developer(s)icCube software Sarl
Stable release
8.4.6 / November 1, 2023 (2023-11-01)
Operating systemCross-platform (JVM)
TypeEmbedded analytics
Websitewww.iccube.com

icCube is known for its embeddable data analytics and visualization software platform tailored specifically for B2B Software-as-a-Service (SaaS) applications, i.e. Embedded analytics.

Its customers serve various industries, from finance and healthcare to e-commerce and logistics among many others. The software enables SaaS solutions from multiple sectors to provide data analytics, dashboards and visualization to their respective end-customers (i.e., Customer-facing analytics).

History

icCube was founded in 2010 by David Alvarez-Debrot and Marc Polizzi, recognizing the need for an analytic server that could be seamlessly integrated into third-party products.

David and Marc, who had previously worked together in developing a financial risk product at a software consulting firm, noticed the necessity for a robust, performant and reliable server for analytic calculations. After the roll-out of the financial risk product, they decided to start their own venture.

Initially considering developing an alternative financial risk solution, they later realized the potential in the market for a versatile, cross-solution product focused on analytic modeling and processing that could be easily embedded into any SaaS solution. Many B2B SaaS companies were seeking robust analytics capabilities for their platforms, and the idea behind the company was to be thought-leaders in this space by focusing exclusively on providing an embeddable analytics platform for B2B SaaS solution developers.

The technology is Java-based, ensuring compatibility with most architectures. The in-memory server uses the Multidimensional Expressions (MDX) query language, which in contrast to other common query languages, is highly optimized for analytics.

Over time, the platform evolved and introduced new features and enhancements to meet the expanding needs of its customers. Noteworthy milestones in the evolution of icCube include the introduction of the Web Reporting server in 2012, the launch of a new reporting system and server calculation engine in 2016, and the release of a new dashboard module based on TypeScript, React, Redux, and Material UI (MUI) in 2022.

DateVersionEvent
June 20100.9.2The very first published version (preview) of the in-memory OLAP server; MDX/XMLA support are the primary objectives.
November 20101The first features complete. A community (free) version.
June 20111.3Expanded MDX support and stronger cube modeling features; the first version of the visualization library (GVI).
October 20112First version advertised for business use (vs. community).
April 20122.5First version featuring the Web Reporting server.
January 20133Better performance and more features.
June 20134A second generation calculation engine.
January 20154.8.2Improving the 4.x versions (server features and speed, Web Reporting).
May 20155.1Adding ETL features.
May 20165.2Improving the 5.x versions.
October 20166.0Brand new reporting and new server calculation engine.
July 20176.2Added Google Maps layers for GEO widgets, heat maps, etc.
August 20176.5Added dashboard commenting module for collaboration.
April 20186.6Improved ETL.
April 20197.0New Server UI / New JSON Rest API.
January 20207.1Support for Java 11 and onwards.
March 20238.0New dashboard module (React, Redux, MUI).
February 20238.4Java 17, multiprocess support for DOCS, available as a Docker[1] libraries in Github.[2]
November 20238.4.6Improved performances and print server.

Technology

Architecture

The product is a fully browser-based application, with the server implemented in the Java programming language (Java 17) following J2EE standards. For the latter, it embeds both an HTTP server (Jetty) and a servlet container to handle all communication tasks. Reporting is developed in TypeScript / React / Redux.

Being an in-memory server, the server does not need to source its data from a RDBMS; in fact, any data source that exposes its data in a tabular form can be used; several plugins exist for accessing files, HTTP stream, etc. Accessing datasource that expose JSON objects is also supported (e.g., MongoDB). The platform then takes care of possibly complex relations (e.g., Many-to-many) implied by the JSON structure.

icCube uses Multidimensional Expressions (MDX) as its query language and several extensions [3] to the original language : function declarations,[4] vector (even at measures level), matrix, objects, Java and R integrations.[5] icCube patented an MDX debugger.[6]

Accessing the platform (data modeling, server monitoring, MDX queries, dashboards) is performed through a Web interface and a JSON REST API.[7]

Running icCube

The product can be self deployed on premises and/or in the cloud. It is available as a Windows installer, Linux ZIP file, Mac version and as a Docker container deployed at Docker Hub.[1]

Key Components

Server

  • Data modeling – the Builder allows for creating data models (schemas) using a Web based user interface. All the requisite parts of a schema: data sources, dimensions, hierarchies, calculated members, etc. can be defined there.
  • API authentication & authorization (SSO) – icCube is hosting a J2EE servlet handler (Jetty) to handle all communications with the server. Authentication is configured via dedicated servlet filters and an internal authentication service. Both can be changed via dedicated Java plugin(s) to meet corporate policies (e.g., SSO). As examples, some customers use Windows SSO, encrypted Web token, etc. Authorization is achieved using roles granting access to resources (i.e., monitoring, schema builders, schema data). Roles can be defined within icCube or generated on-the-fly when users are connecting to the platform using dedicated information (e.g., a list of schemas the users have access to).
  • Management API – A REST API (i.e., JSON over HTTP requests) is provided for managing the server and the available schemas.

Dashboards

The Dashboards application allows for creating web-based dashboards, based on widgets, gadgets and events:

  • Widgets are visual elements such as graphs, charts, maps, filters, etc.
  • Gadgets are preconfigured, reusable widgets.
  • Events manage interactions between widgets. For example, a filter generates an event that produces an action on a chart. Also, widgets can also contain events and have actions on each other.

Embedding icCube

The icCube Dashboards API[8] allows for:

  • developing new themes, widgets (charts, maps, etc.), data transformations, etc
    • Plugin dev kit[9]
    • Github: ic3-reporting-api[10]
    • Source code examples[11]
  • embedding icCube Dashboards into a Web application
    • Embedded API[12]
    • Github: ic3-reporting-api-embedded[13]

See also

References

  1. 1 2 "Docker".
  2. "Github".
  3. "Chris Webb on icCube MDX Declared Functions".
  4. "MDX Functions Reference (common + extensions)".
  5. "icCube extends MDX with OO capabilities".
  6. "Debugging system for multidimensional database query expressions on a processing server".
  7. "JSON REST API".
  8. "icCube Dashboards API".
  9. "Plugin dev kit".
  10. "Github: ic3-reporting-api".
  11. "Source code examples".
  12. "Embedded API".
  13. "Github: ic3-reporting-api-embedded".
This article is issued from Wikipedia. The text is licensed under Creative Commons - Attribution - Sharealike. Additional terms may apply for the media files.