GeoBI Architecture Based on Free Software: Experience and Review - Geographical Information Systems: Trends and Technologies - page 278

the previous section. For example, Bimonte et al. (2005) used JPivot to
provide JSP pages with OLAP functionality and MapXtreme Java to support
map visualization and interaction. On the other hand, Silva, Times et al.
(2008) based their SOLAP front tool on the Java Plugin Framework (JPF),
OpenJUMP, and JRubik technologies. These SOLAP clients can be also
integrated with Google Earth providing 2D (Silva, Manhäes and Gitahy
2008) or 3D (Di Martino et al. 2009) view of spatial data. These systems differ
in the presentation of multidimensional elements and also in functionalities
since the authors' goal is to demonstrate feasibility of implementation of
the concepts proposed by them.
There are various OLAP front-end solutions that use Mondrian as an
OLAP server providing basic OLAP functionalities, e.g., JPivot, STPivot,
JPalo. However, implementers of SOLAP systems based on a GeoMondrian
server have a diffi cult time fi nding free SOLAP front-end software. The
SOLAPLayers (Spatialytics 2013c) tool was fi rst announced to be released
together with GeoMondrian; however, it is no longer supported and a new
GeoBIExt will be developed to replace it. To the best of our knowledge,
this extension is still unavailable. On the other hand, some initiatives are
currently emerging with the aim to join their effort in order to develop
fully-functional SOLAP software. For example, the SpagoBI initiative
(SpagoBI 2010) proposed to create a group of developers to take advantage
of existing solutions, e.g., GeoOLAP from Camptocamp or GeoMondrian
from Spatialytics. Currently, SpagoBI includes a component called Location
Intelligence (SpagoBI 2013a) that combines GIS visualization and spatial
analysis with BI features related to conventional data. However, this
extension is relatively new and still requires manual confi guration using
a new document composition internal engine (SpagoBI 2013b) in order to
be able to synchronize tables, graphs, and maps.
Another solution is Saiku 2 (Saiku 2013) which allows the selection
of cubes, dimensions, measures, and typical OLAP operations with the
connection to Mondrian, SSAS, SAP BW and Oracle Hyperion. Although
Saiku does not include spatial support natively, Lamas et al. (2013) propose
the extension based on a map server developed to deliver maps. However,
this extension again is not freely available and required LeafLet, an open-
source JavaScript library for mobile-friendly interactive maps, in order to
integrate OLAP cubes and maps. Furthermore, a new tendency has also
emerged that relies on Web services as providers for SOLAP environments
(Barros and Fidalgo 2010).
We believe that the front-end tools as a free option are currently the
weakest component considering different layers forming a SOLAP system.
The lack of this layer introduces an additional GIS-like level complexity
2 Previously called Pentaho Analysis Tool and currently in development to replace Pantaho
JPivot.