 "The DLM Strategy Blueprint")
Data Lifecycle Management (DLM) and Information Lifecycle Management (ILM) are terms that are often used interchangeably. Can you highlight the distinction between the two?
I agree that DLM and ILM are words that are often used interchangeably. However, I believe that there is a vast difference between the two. ILM is a super set and DLM is its sub set. DLM deals with physical attributes of data such as file name, age, date, etc. whereas ILM deals with the information that resides inside that data.
To cite an example, a company can define a policy with help of DLM that can say a certain type of data can be archived after a given number of years and deleted after the next few years. In ILM, a company can know what is in the data. In our case, it could be customer account number which can then stored and archived according to its relevance. ILM has an intelligence built in it that can help create policies to manage data.
DLM is not just a solution; it is a comprehensive approach towards managing a company’s data. Please elaborate on how you have incorporated this approach in your organisation. What are the business benefits that you have reaped?
The objective of DLM is to store and archive data in such a way the information that is most relevant and critical is stored in high speed expensive disks. Data that is older is then moved to less expensive storage modules. DLM also helps in easy retrieval of data as and when required.
For us, the objective of DLM is to archive customer transaction information. All the transaction information is on current tables that are stored on Solid State Drives. These drives are extremely high speed and ensure access is fast and in real-time. Data is stored in this drive for two years. After this, data is then moved to the archives table for the period of six years. The need to access this data is on a year-on-year basis and we can afford to put it on a slower media. Post this duration, data is then moved to backup on an offline media i.e. tape. Data then resides outside the system.
This approach has helped us drastically cut cost of storage, increase speed of transaction, boost efficiency and manage data effectively.
Do you believe that DLM can help organisations meet compliance requirements?
Company policies and regulators such as RBI have certain rules that indicate the life of information. It is important to note that when most of these laws were framed, information was still in the physical form. Now even after information has been digitised, same laws are still applicable. Compliance is forcing organisations to retain data for a certain period of years. DLM plays a major role here as it helps in the entire process of data management and archival according to relevance of information.
Can DLM address the challenge of unstructured data within an enterprise? Please elaborate.
Almost 60 percent of data in organisations is unstructured data. To asses what this data is and where it lies is a daunting task. IT alone cannot undertake this mammoth exercise. Firstly, there needs to be top management buy-in. It is important the whole organisation participates in it.
Classifying data is easier in organisations where majority of customer data is in the form of invoices and bills. It gets difficult in companies where there are word documents, excel sheets, etc. created to support customer information.
Yes, there are solutions available that can help make sense of unstructured data. DLM and ILM can be very effective here. However, effectiveness of these tools depends on how they are utilised by the users.
What are some of the key best practices that should be followed for an effective DLM implementation?
According to me, best practices that need to be followed are:
Create an inventory of all company data
Clearly define policies that will determine how long to retain different types of data based on compliance as well as organisational requirements
Ensure there is top management commitment
Select the right tools
- Ensure periodical monitoring of the DLM strategy