Pentaho BioMe App Demo Installation, Access & Usage Instructions
Pentaho BioMe App DemoInstallation, Access & Usage Instructions
Project SDR Target Pentaho 5.4 Version 1.0 Date Jun 2015
Content
1. Overview of BioMe App Demo
. Technology and Access Overview
2. Local Installation
. Minimum Requirements
. Local Installation Instructions
3. Usage and Navigation Instructions
. Introduction to Application Usage and Navigation
. Walkthrough Completing and Visualizing a Dataset Request
. Cleaning up Dataset Request
. Viewing of BioMe App Demo in Multiple Devices
Overview BioMe App Demo
1.
Splash screen upon the starting of the BioMe App demo
that contains related information.
FIGURE 1
Technology and Access Overview
The BioMe App demo is to be accessed in two ways:
. Locally installed and running demo.
. Hosted demo site that is remotely running and requires no local
installation.
The BioMe App demo was developed to be a standalone web application
using AngularJS and to not be required to be accessed via the Pentaho User
Console (PUC), as was the Movie Lens and GartnerCensus SDR Samples.
Although, it is still required that the user of this demo to first be logged into
PUC (for local installs), due to the BioMe App demo’s HTML being served up
and authenticated by PUC.
Locally Installed Hosted Demo Site
The technologies used to build the BioMe App demo were:
Out of the Box
. Pentaho Data Integration
to design and automate the data set request process behind the scenes
and streamline data refinement.
. Pentaho Analyzer
for pivot-style ad hoc analysis and visualization of data.
. Pentaho CTools
for the custom dashboard in the main view as well as the data request form.
Customizations
. AngularJS
Pentaho components were embedded into a 3rd party application that
leverages AngularJS for dynamic views.
. Pentaho’s JavaScript APIs
were leveraged to embed Analyzer in the 3rd party application and power
pre-populated chart buttons and timeframe controls.
. RequireJS
a JavaScript file and module loader, was leveraged in embedding the CTools
dashboard into this application.
. Bootstrap
The BioMe embedded dashboards were built with Bootstrap, a responsive
front-end development framework.
The components that make up the SDR are PDI for parameter entry
and an app for refining the data. This app calls to the PDI backend of the BA
server for the main job: refining data through PDI using the new Build Model
job entry, then publishing the data source back to the Business Analytics (BA)
Server through the Publish Model job entry. Once it is published, the refined
data is available for use in viewing the canned Analyzer reports.
Local InstallationLocal Installation2.
Minimum Requirements
To complete a local installation, the following are required to be installed:
. Pentaho 5.4 Platform downloadable installation using default settings.
. BioMe App demo downloadable installation using documented
instructions.
Browser compatibility:
The BioMe App was designed to run on recent versions of browsers
(Firefox, Chrome, Safari), but was not widely tested.
Firefox Chrome Safari
Default installation method using Postgres: choose
default for installation, and then enter password for
the Postgres password information.
Ensure the use of port 8080 for the BA server.
Default authentication is required.
Install Pentaho Software
Download and install
Pentaho-business-analytics-5.4.0.0 software for your
environment: Windows (exe), OS X (tar), Unix (bin).
The following are installation defaults that are required:
Log in to PUC,
and verify that it is working properly.
.
.
.
1
2
Extract
the contents of the Zip file to this directory:
pentaho/server/biserver-ee/pentaho-solutions/system.
It should create a BioMe folder in the directory.
Open up
web browser and open up the following URL to start
the BioMe SDR 5.4 Demo app:
http://localhost:8080/pentaho/api/repos/BioMe/
webapp/index.html
Download and Install the Demo
Download
the BioMe App Demo Zip file from the following Box
folder, BioMe App Demo Content.
1
2
3
Usage & Navigation Instructions
3.
Introduction to Application Usage & Navigation
After clicking on the splash screen’s Proceed button - FIGURE 1, the BioMe
App Demo will be viewable and will display navigation/selection panel,
dashboards, static and hardcoded Key Performance Indicators (KPI), and
navigation to the Dataset Request form as shown in FIGURE 2. Scroll down
as to see the remainder of the Country Card dashboard values.
1. Static KPIs.
2. Left vertical selection panel: Only analytics is active.
3. Single dashboard with time analysis and country card views (CTools).
4. Navigation to Dataset Request Form.
FIG. 3
1
2
3
3
4
OVERVIEW
1
2
3
3
4
Within dashboard’s Time Analysis view, you can hover the mouse over the
2015 World Avg. line, colored in black, as to see a hotspot value, shown in
FIGURE 3. You can also hover over the top of the per year values, colored in
green, as shown in FIGURE 4 and a hotspot value will be displayed.
You can view the differences between the two values in this manner.
OVERVIEW
FIGURES 3 / 4
The dataset request form has the following components, FIGURE 5, to help
complete and utilize dataset requests:
1. Request List that stores dataset requests by data source name and allows
for viewing status, duration, and to be selected (by clicking on one that has
completed its processing) to be viewed in analyzer reports.
2. Filters to reduce narrow the search and result set.
3. Data Source Name that will annotate the data request for processing
status and reuse in the Request list.
4. Analyze button to be clicked after filters have completed and data source
name has been entered, which will then run the SDR.
1
2
43
OVERVIEW
FIGURE 5
Walkthrough Completing & Visualizing a Dataset Request
Now let’s walk through completing a Dataset Request. In the Country filter,
select 10 countries (Austria, Netherlands, New Zealand, Norway, Panama,
Poland, Portugal, United Arab Emirates, United Kingdom, and United States of
America) and then in the Years filer, select the values of 2005-2010 as shown
in FIGURE 6. Enter a data source name of “Country Year Filter” and then click
on the Analyze button and one of the windows in FIGURE 7 will be displayed
(due to duration of processing dataset request) that will allow you to go to
Analyzer or back to form and await completion of the dataset request.
OVERVIEW
FIGURE 6
Note: No ability to clear one/all filters or to select/deselect all values in a single filter is available at this time.
Clicking on the Go to analyzer button will take to FIGURE 9.
Clicking on the Close or Notify when ready buttons, will take you back the
application’s main screen where you will have to click on the Show Dataset
Request link on the bottom middle part of the window to be taken back to
the Dataset request form shown in FIGURE 8.
Redirected to Analyzer
Click on the Latest Requests list and then click on Country Year Filter request and you will be redirected to FIGURE 10
FIGURES 7 / 8
The embedded analyzer, shown in FIGURE 10 has the following
components:
1. Navigation Back to form bottom that will take to back to the Dataset
Request form.
2. Canned Report Selector to choose between three canned reports of a
chart, column bar and scatter plot.
3. Analyzer Menu Bar to view (Edit Mode) or hide (View Mode) panes,
view/add filters, and perform other actions.
4. Report Type Selector to change the report type visible within the
embedded analyzer.
3 4
1 2
OVERVIEW
FIGURE 9
The default canned report that will be loaded is a chart report within an
embedded analyzer in view mode. Select on Column chart (middle
selection of canned report selector) and FIGURE 10 will display a column chart
within the embedded analyzer view mode. Time Frame Slider is available as
to filter data.
OVERVIEW
FIGURE 10
Can change from a column to bar report view within the embedded
analyzer with the Report Type selector shown in FIGURE 11.
OVERVIEW
FIGURE 11
Select on Scatter Plot report in the Canned Report selection area and
FIGURE 12 will be shown, which has a Year of Analysis Slider to filter data.
Plotted data can then be selected using the mouse and creating a box over
the scatter plots of concern, as shown in FIGURE 12. In this example, we have
select the bulk of the scatter plots, and have left out 2 outlier to the left and
3 outliers to the right.
OVERVIEW
Use mouse to create box over scatter pointsFIGURE 12
After you release the mouse button, Keep Only/Discard/Clear selections will
appear as shown in FIGURE 13 that will allow for further analysis.
Select Keep Only. A refreshed Scatter Plot will be displayed and further
analysis can be performed by this view, changing it to a grid view, etc. using
the Report Type Selector.
OVERVIEW
FIGURE 13
Click on the left selection of the Canned Report Selector, as to view the
canned grid report. The embedded analyzer canned grid report can be
customized by making visible the fields and layout panes via the analyzer
menu bar shown in FIGURE 14. This will place the embedded analyzer in the
Edit Mode and allow for changing of Row, Column and Measure selections,
shown in FIGURE 15.
OVERVIEW
FIGURES 14 / 15
FIGURE 16 shows a revised grid chart using different column, row and
measure selections. This was accomplished by removing the pre-existing
selections in the Layout pane and then dragging and dropping the new
selections from the Field pane in their according column, row and
measure layout pane areas.
FIGURE 16
Cleaning up Dataset Requests
The ability to delete Dataset Requests completed via the Dataset Request
form is available, shown in FIGURE 17. Click on the Request List on the
Dataset Request form. Once the list is displayed, click on the Edit and then
FIGURE 18 will be displayed. Click the checkbox next to the Country Years
Filter Dataset Request and then click Delete. FIGURE 19 will be displayed
and you will have the choice to continue, by clicking OK button, or cancel
the delete request.
Request List
Click to see drop-down list of requests.
Click edit button to delete a request.
1.
2.
Delete Request
Click checkbox then, click Delete.
3.
A popup will be displayed and you will have the choice to continue,
by clicking OK button, or cancel the delete request.
FIGURES 17 / 18
Viewing of BioMe App Demo in Multiple Devices
The BioMe App Demo can be viewed across multiple devices, laptops, tablets
and phones. This was achieved through the use of Bootstrap and standard
HTML5 + CSS3 features.
OVERVIEW
Tablet PhonesLaptops
Launch Demo
Interactive BioMe Demo