Data Fragmentation occurs when data is broken down into smaller pieces and stored in different locations. Data can be physically fragmented, where data is spread across different physical devices or servers. Data can also be logically fragmented where a single file can be broken into multiple parts and stored across a storage system. Aside from that, there can be fragmentation on the application level where applications may store the same data in their own formats.
The easiest and simplest way to handle this challenge is to implement a centralized platform. This single source of truth can aggregate, process and store data from multiple sources. But this solution is hard to implement considering the abundance of applications that enterprises deal with today. The application sprawl at enterprises means that one repository may not be sufficient to handle the diversity of data that is coming into the enterprise every day. Enterprises can also draw up a comprehensive data management strategy to organize data infrastructure, optimize usage, consolidate data as much as possible, and ensure continuous vigilance to combat data growing in silos furthering data fragmentation.
How does fragmented data affect unstructured data?
Fragmentation has its costs and they multiply in several ways when the data is unstructured. For instance, unstructured data such as images, videos, sound clips, large blueprints etc. can be duplicated across multiple data silos, which can lead to storage costs. Unmanaged data fragmentation can slow down data processing and affect business operations. For example, in healthcare, data fragmentation can lead to misclassification errors in algorithms. Fragmented unstructured data can escape data security and governance strategies, increasing the risk of data breaches. It can complicate eDiscovery, increasing litigation costs.
All of this can happen before the cost of productivity where employees can spend more time searching for information across apps and storage repositories.
Comments