Application migration is one of the most common phenomena in PLM applications as company’s moves from legacy application to newer applications as per their changing business needs; however their need to share data with suppliers and customers remains the same.
The challenges to migrate the data are huge from one application to another are huge:
a) Data needs to be reformatted as per target (new) application need.
b) Data cleansing is required.
c) New business rules needs to be written as per new business requirements.
Approaches to support data migration
ETL data migration approach: This is so called one of the best practice methodologies; it involves using various tools for individual parts of data migration. However it is not a holistic approach that maximizes the business value of the information that is migrated to the source PLM systems.
Customized Data Migration Strategy: It involves using various practices to migrate data from source system to target system depending on the amount of data that needs to be migrated plus the time constraint in which data needs to be migrated
Migration Strategy
• Data Profiling: Data present in legacy systems needs to be checked before it is moved to source system as data may be redundant, similarities and discrepancies.
• Understand Data: Need to understand the complete structure of the data.
• Review of dependable data sets: Review what data sets are dependent on other data sets.
• Data Cleansing: Data needs to be cleansed to remove redundant, similarities and discrepancies.
• Select a Migration Strategy: There are two approaches that be used here:
o Migrate complete data at one go: Complete chunk of data is taken from source system and is migrated to the target system and source (legacy) system is completely shut down.
o Migrating Data in Chunks: This is often known as phased migration approach in which data from legacy application is migrated to target application in various phases, legacy systems are not shut down completely.
• Maintenance of Data: Original data needs to be maintained till the validity of the migrated data is tested properly, need to run the legacy and source system in parallel for some time.
• Test the Migration Process: Test the migration process on small chunk of data that is migrated. To get information about how time it will take to test the complete migrated data.
• Train the Users: Users needs to be train regarding how the data is present in the new system, so that they can use the data once the migration is completed
• Perform Data migration: Migration needs to be done in the most disruptive manner. If the system allows co-existence then migration can be performed in chunks without disturbing the business.
Among the several data migration consulting companies , I found your company to be one of the best, which helped me in designing the proper plan for transforming the data correctly.
ReplyDelete