Data migration is the process of transferring of data between storage types, formats, or computer systems. An automated migration is achieved by developing programs that performs the data migration and frees up human beings from tedious tasks.
Definition : Data migration is defined as the process of transferring data from one repository to another.
Another definition : The periodic transfer of data from one hardware or software configuration to another, or from one generation of computer technology to a subsequent generation is called data migration.
Data migration is a fairly time–consuming process but the benefits are worth then this cost. The other major advantage of data migration is that the old applications need not to be maintained.
1. Basic Concepts of Data Migration
Some of the basic terms that are needed to understand the data migration is as follows:
Legacy Data : It is defined as the recorded information that exists in the current storage system. This recorded information can include records in the database, text files, images and spread sheets. This recorded information can be migrated to a new system.
Data Cleansing : It is the process of preparing legacy data for migration to the new system. The data cleansing process involves manipulation or cleaning the legacy data so that it conforms to the new system’s requirement.
2. Why We Need Data Migration
Data migration is generally required when organizations change computer systems or upgrade their old systems to new systems. The legacy data stored on out of date or obsolete formats is evaluated, indexed, deduplicated and then migrated to newer more cost-efficient and a reliable storage media. Data migration is performed programmatically so that an automated migration can be achieved. The various reasons why we need data migration are as follows :
- The first and the major reason is the natural death of the IT Every IT system one day reaches the end of its life cycle and has to be replaced.
- It is possible that the hardware is still functional but there may be following symptoms of decease like
- Huge maintainance costs.
- Loss of connectivity e., networking problems.
- Storage bottlenecks.
- User bottlenecks.
- Unbearable performance.
These all factors in combine or individually force data migration.
- The third major reason is the customer’s informational These include WWW connectivity and integration of other data sources that require new IT systems and refurbishment of data structures.
- The fourth reason may be the implementation of a new data processing paradigm for the whole company or control data base servers.
3. Key Factors in Data Migration
Once we have decided for the data migration, various factors need to be examined to decide whether the whole dataset or some part of the dataset or none of the dataset should be moved over to the new system. The two key factors that must be considered in deciding the data migration are Data Volume and Data Value.
Data Volume : It is the easiest factor to evaluate in the decision process. The major questions to be answered in this are:
- How many data records are there?
- How many are expected to come into new system on weekly or monthly basis?
- Is there any technical road blocks to bringing over a certain amount of data?
- Whether large databases will affect the performance of the system function?
We have to find the answers of the above questions and then decide accordingly. If the volume in low, then performing migration is worthful as we have some database for the users and for trend analysis.
Data Value : It is much more harder factor to evaluate than the data volume. Many times various perceptions exists concerning what value the existing data will provide. If users have no working experience with older data in the current system, it is possible that they may not work with older data in the new system as well with the improved functionality. So, we have to look at short-term parameters. Thus, extracting the exact data based on the various factors will depend on the abilities of the current system and database as well as the ability to write the detailed extraction script.
4. Stages of Data Migration Process
A well designed migration strategy ensures that the data is properly evaluated, reviewed and restored before it is migrated to new, more accessible and cost-effective media. A proper strategy makes certain that the valuable data of the organisation is safeguarded from beginning to end. There are mainly four stages of a data migration strategy as shown in Figure 15.1.
- Evaluation of legacy media.
- Review of customer requirements.
- Restoration of data.
- Actual migration.
- Evaluation of legacy media : The evaluation of the legacy media is performed in this The media may be of any brand or any type.
- Review of customer requirements : The customer requirements are evaluated in this The major considerations are whether the data is being prepared for business, compliance or litigation requirements. In this stage, the targeted media like disc or other formats is also evaluated and determined.
- Restoration of data : In this stage, the data placed on different media are read and the necessary data in recovered before the data extraction process Also, the individual files are identified and restored.
- Actual migration : The data that is restored in the previous stage is indexed and deduplicated in this stage. Further, unnecessary and redundant files are purged to reduce the storage requirements. Now, the data is ready to be migrated to the new storage media and is migrated. After migration it is delivered to the customer.
Source: Gupta Satinder Bal, Mittal Aditya (2017), Introduction to Basic Database Management System, 2nd Edition-University Science Press (2017)