IBM InfoSphere Master Data Management Silent Installation

In a silent installation, the IBM Installation Manager is invoked passing in a file (called the response file) containing parameter values required for installing Master Data Management.  This includes the Install Location, the location at which the MDM installable is available, the location at which the Shared resources can be stored, the features to be installed and configuration parameter value for each feature.

Sample response files for installing MDM are available in the MDM Startup Toolkit.  The Installation Manager can also be of help in creating a response file.  The record option can be used and a file name can be specified.  The Installation Manager operations will be recorded and the parameter values provided during installatin will be stored in the response file.  This file can be used as is, or with required modifications.

Recording from Installation Manager to create response file for Silent Install
C:\Program Files (x86)\IBM\Installation Manager\eclipse>IBMIM.exe -dataLocation D:/InstanceLogs/MDMLogs -record MDMInput.res

Silent install using a response file
C:\Program Files (x86)\IBM\Installation Manager\eclipse>IBMIM.exe -dataLocation D:/InstanceLogs/MDMLogs -silent -input MDMInput.res -acceptLicense -nosplash

The Silent installation process is quicker, although the validations provided by the Installation Manager GUI are not available with it.  Silent installation response files can be generated and used when installation has to be performed multiple times. 

The most important advantage of Silent install is that it allows the user to provide customized values for certain inputs, which is not possible with the Installation Manager GUI.  For example, when a custom WebSphere Application Server profile trust file is created, the path of this file and the password for this file cannot be provided when Installation Manager GUI is used.  However, these values can be modified and specified when Silent installation is used, using the below parameters.

<data key='user.L2.was.node.client.trust.file.url,com.ibm.mdm.advanced' value='etc/DummyClientTrustFile.jks'/>
<data key='user.L2.was.truststore.pwd,com.ibm.mdm.advanced' value='WebAS'/>

Key parts of the response file
variables 
  <variables>
    <variable name='sharedLocation' value='D:/Instances/Shared'/>
  </variables>
Declares the Shared Location used by the Installation Manager

server
  <server>
    <repository location='D:\Installables\MDM\Cuba\MDM\disk1'/>
  </server>
Holds the details of the MDM repository

profile
  <profile id='InfoSphere MDM Standard Edition or Advanced Edition' installLocation='D:\Instances\MDM114'>
    <data key='eclipseLocation' value='D:/Instances/MDM114'/>
...
</profile>
Holds the profile id, the install location and the entire set for parameters for configuring the instance

install
  <install modify='false'>
    <!-- InfoSphere MDM Standard Edition or Advanced Edition  11.4.0.0 -->
    <offering profile='InfoSphere MDM Standard Edition or Advanced Edition' id='com.ibm.mdm.advanced' version='11.4.0.FP00IF000_20140924-0832' features='com.ibm.im.mdm.db.feature,com.ibm.im.mdm.app.feature,com.ibm.mdm.ba.webapp.feature,com.ibm.mdm.ev.webapp.feature' installFixes='none'/>
  </install>
Holds the details of the offering, the id, the version and the features to be installed.  It also indicates whether the installation is a new one, or it is a modification of an existing instance, using the flag modify.

uninstall
Indicates that the instance has to be uninstalled.

preferences
A set of preferences are mentioned.

Links:
Silent installations using response files
Installation Manager command line arguments

Separation of Installation and Configuration - InfoSphere Master Data Management Installation

Installation of InfoSphere Master Data Management involves two major steps:
1. Placing all the components required for configuring the Master Data Management instance in a given directory (after performing required checks)
2. Configuring the application server, database and other services based on properties provided by the user.

Starting with InfoSphere Master Data Managment version 11.3, there is an option provided to configure the MDM instance after installation.  This option is also known as Separation.  When Separation is used, features selected for installation and the properties provided by the user are stored in certain files.  Configuration of the instance is performed based on values read from these files.

To make use of this feature, the option Manually run the scripts to configure InfoSphere MDM components after installation has to be selected in the Install Configuration Review Panel.

Installation Manager UI Summary Panel

When silent install is used, the below property has to be set to true in the input file provided.

<data key="user.L2.separate.install.config,com.ibm.mdm.advanced" value="true" />

After the Installation is complete, a target Configure_MasterDataManagement has to be invoked from the <MDM_INSTALL_DIR>/mds/scripts.  Database and Application Passwords have to be provided to configure the instance.  

Invoking Configure_MasterDataManagement

Further details on Separation and its usage after adding or removing features to MDM (Modify) and Uninstall are explained in detail in Separating Configuration from Installation

Separation is easy to use, as it maintains the properties and requires minimal details when the target is invoked.  It is highly helpful to debug installation failures when there are minor errors in the values provided for configuration.  InfoSphere MDM installation provides logs at the <MDM_INSTALL_DIR>/logs folder, wherein the database and madconfig targets logs are available.  Logs created during installation are also available under a directory which is provided as the dataLocation property to the Installation Manager  There are sufficient checks available to ensure that an installation would not fail.  However, installation failures occur rarely.

When Separation is not enabled and an installation fails,  debug has to be done based on the files in the logs folder under <MDM_INSTALL_DIR> and the dataLocation parameter provided to the Installation Manager.  The issue has to debugged and the install has to be started from scratch.

When Separation is used, files required for configuring the instance are available in the <MDM_INSTALL_DIR>  Configuration of the instance (which includes creating virtual tables, physical tables, server configuration, installation of the native ear, eba, webs services, configuring security and configuring selected UIs) occurs when the target ./madconfig.sh Configure_MasterDataManagement is invoked.  When all the parameters provided are valid, the configuration proceeds smoothly and completes.  The targets being executed as part of the configuration and the logs generated are displayed on the console.  The log files generated during the configuration are available under <MDM_INSTALL_DIR>/logs/madconfig directory.  The entire log created during the target execution are also available in the same directory.

Assume that one of the parameters giving during installation is incorrect. Supposing an incorrect value is provided as the database schema name, the configuration will fail when an attempt is made to create the tables.  After the Virtual tables are created, a check is performed by the Configure_MasterDataManagement to ensure that the tables are present.  This check would fail and indicate the user about the absence of tables.  Although a failure occurred , all the files would remain in the <MDM_INSTALL_DIR>.  It becomes easy to figure out where the problem occurred due to the presence of log files.  Also, once the problem is figured out, it could be fixed in the properties file(s) that are present in the <MDM_INSTALL_DIR>/properties directory and the Configure_MasterDataManagement target can be invoked again to configure the instance.  Thereby, Separation of installation and configuration helps us save a lot of time. 

The target maintains details on the status of all features, whether a feature is configured, partially configured or not configured.  When a feature is not configured, and is part of the list of features to be available, the feature will be configured.  When a feature is partially configured (in general, a feature is partially configured when a failure occurs during its configuration), an uninstall of the feature is performed followed by the configuration.  Similarly, removal of configured features is performed for features that are not required (this case generally happens during a Modify or Uninstall).

The Uninstall_MDM target has to be invoked before performing an uninstall from the Installation Manager.  Also, it is possible to invoke Uninstall_MDM followed by Configure_MasterDataManagement in case of any necessity.  In case any additional requirements are identified after an installation and configuration, the above option could be used.

InfoSphere Reference Data Management can also be installed using Separation and the target Configure_ReferenceDataManagement can be used for the configuration. 

A stitch in time saves nine - How do I find the time? Explore and Derive Insights using InfoSphere MDM and Watson Explorer

Assume that you are a Customer Representative working for a broadband company.  Your company has installed IBM InfoSphere Master Data Management and you have a trusted view of all your customers.

You keep calling each customer regularly, requesting whether they want an increase in the quantity of data with a good speed, but hardly few of them agree.

How about having a system which will alert you of customers who go very close to the quantity of data they have subscribed for, and go more than that for a couple of times.  Wont they quickly agree and move to your plan?

Now how do you get to them?  The best way is to integrate the IBM MDM with IBM Watson Explorer.  With IBM MDM giving the trusted view of customers and Watson Explorer interacting with other sources to find more details about your customers, it would become easy to get a complete 360 degree view of your customers.  The analytical details provided by this integration would help you find the right time to convince the customers who require such a broadband plan.

The benefits that can be reaped by the IBM MDM integration with IBM Watson Explorer are many and details on the advantages and integration are clearly described in the below links.
Exploring Data Insights with IBM InfoSphere MDM and Watson Explorer
Access Data Patterns with IBM Watson Explorer
Building 360 degree Information Applications

Master Data is within systems in my organization, how do I integrate this with a CRM on Cloud

IBM InfoSphere Master Data Management and IBM WebSphere Cast Iron Cloud Integration come to help you here.

IBM Master Data Management provides your a single version of truth of your entities.  In essence, it helps you cleanse data and remove duplicate values, thereby maintaining a single record of the entities in your system.  Typically master data requires more privacy and is maintained in systems in your premises.

There are other software that are available over the internet as services (SaaS on the cloud).  There will be requirements wherein master data in the organization's premises, have to be used with data in cloud.  There has to be means through which this integration can take place. Data security is a key point of concern in such integrations.

IBM WebSphere Cast Iron can serve as a middleware between IBM MDM and SaaS applications.  There are proven implementations of IBM MDM with Salesforce.com, Netsuite, Workday and IBM Bluemix services using the IBM WebSphere Cast Iron Cloud Integration.

Links:
IBM InfoSphere MDM integration with SaaS applications using WebSphere Cast Iron Cloud
Master Data Management for SaaS Applications
Putting master data into Salesforce.com

IBM Watson Analytics

IBM Watson Analytics helps you to explore data and to discover predictive insights hidden in your data in minutes time.  A Freemium version is available and we can identify the advantages it provides in a short duration.

Explore, Predict and Assemble are three primary features available through IBM Watson Analytics.  An initial data load is required before using these features.  There are tutorials available after login, which provide us an idea of how these features can be used.

Apart from the tutorials, a tour comes with each feature, which explains the steps to be used to gain value from the feature.

IBM Watson Analytics is a boon for enterprises.

Below are steps I followed to try IBM Watson Analytics:
1. Created a csv file with fields Month, City, Equipment, Sales, Discount and Festival, having 33 rows, and Added this file using Add File option.

2. Used the Explore feature and followed the steps that come as a tour to generate the below.

Explore

3. Used the Predict feature and followed the steps that come as a tour to generate the below.

Predict

4. Used the Assemble feature and followed the steps that come as a tour to generate the below.

Assemble

With more features, tutorials and examples, please explore IBM Watson Analytics.  IBM Watson Analytics Freemium is available now.