Data migration without downtime | Pipeline Magazine

By: Bernd Pruessing

Data migration – the process of preparing, extracting and transforming data and moving it from one system platform to another is difficult. To research shows that 40% of data migration projects go over time, over budget, or fail altogether. Data migrations have become even more complex in recent years due to new technologies, such as virtualized functions in the data center, and new use cases like automation. This makes it even more difficult to extract data from the source, transform it and load it into the new target system.

For service providers and telecommunications operators, management systems are particularly complicated and sensitive to the common challenges of data migration. After all, their networks have complex data models that encompass many different technologies from various vendors. However, if the correct framework is followed, a system replacement without downtime is achievable.

First, let’s explain why it is necessary to migrate data in the first place.

Network inventory management systems are essential for keeping track of IT and network assets. They let you know which devices are on the network, what their configurations are and how they are interconnected, when software licenses will expire and whether assets are reaching the end of their life. They also allow the planning of new equipment for new customers, the planning of changes for network optimization and the dismantling of equipment when customers leave.

Every telecommunications company will, at some point, have to consolidate its inventory systems or replace an existing inventory management solution. One of the common reasons is to tackle new technological challenges such as 5G deployments or hybrid networks based on VNF. These technologies have requirements that many current management systems cannot handle. Virtual network functions, 5G optical networks, FTTx, and flex-grid all require more advanced methods than most older systems can provide.

A strategic goal for many operators is to adapt to new comprehensive network planning and automation use cases, which may not be possible with existing tools. Manual documentation, which includes Excel, AutoCAD, and wiring diagrams, is still in use but is not compatible with the growing dynamics of modern networks and the end goal of network automation. Another frequently cited reason is the elimination of fragmented system landscapes that have built up over time. Often, these fragmented solutions operate in silos, with a combination of proprietary databases and specific inventory management for the data center IP network and DWDM network. Having several parallel systems like this is a drag on efficiency.

Whatever the trigger, inventory management system data is an asset that must be properly documented. This data is the single source of truth for the network and supports critical business functions such as asset management, capacity management, equipment and service planning and deployment, impact analysis, enrichment of alarms and many others. Important decisions are made based on this data, so system availability and accuracy are of the utmost importance. A smooth and uninterrupted data migration and efficient commissioning of the new system is essential to avoid operational damage and business impacts.

Traditional data migration processes have inherent flaws that compromise the success of the migration. These frameworks are based on sequential data dumps and multiple migration iterations and adaptation of migration rules. Every time a migration sequence is performed, the risk of content changes corrupting the final migration is a real threat. This type of migration process also requires system downtime between shutdown

Sean N. Ayres