Master Data Management on Cloud Journey

This article was originally published in DATAVERSITY on July 19, 2021

Most companies are either considering cloud, if not on a cloud adoption journey already. If you are responsible for Master Data Management (MDM) in your company, you are likely considering moving or implementing MDM on the cloud. Although there are still some challenges and kinks to work out in cloud technology, there are cost efficiencies, agility benefits, and many other advantages that should be driving you to look at cloud as a base for your current and future MDM implementations.

When implementing MDM on the cloud, you will need to look at different offerings that combine product capabilities with cloud management service levels. Some vendors offer newer MDM products based on cloud-native architectures that can take advantage of popular cloud Infrastructure-as-a-Service (IaaS) capabilities, while others have not modernized their legacy products and only offer cloud services that cannot leverage some of the more dynamic cloud IaaS features.

If you have not yet implemented an MDM solution, you should look for one that runs on cloud. If you have an existing MDM platform that has been operating for years on-premises (on-prem), you would be wise to look at options for migrating your MDM implementation to the cloud. Having said that, your cloud options for MDM may not be obvious. Let’s explore some of the key concepts to consider as part of your decision.

Infographic

Considering the MDM on Cloud journey? Here is a quick guide for you.

Cloud and/or Managed Service Models

• IaaS: With this model, you purchase IaaS services from a cloud vendor but assume the responsibility for all aspects of the solution above the cloud infrastructure IaaS layer. Account management and segmentation of the application environments, networking, security, install/configure, monitoring, and support for the implemented services must be provided by you as they are in an on-prem environment. The vendor will provide the underlying infrastructure resources, virtualization, some amount of monitoring and/or tooling, and the associated management interfaces for monitoring and managing your cloud infrastructure resources. This would mean that you would be responsible for designing, installing, configuring, testing, and managing your MDM environments running on a vendor-provided cloud IaaS service. This option offers ultimate control and customization but requires you to shoulder all services above the infrastructure layer.
• IaaS with vendor-provided managed services: This option builds upon the first style by offloading selected managed-service tasks to service vendors. Cloud services offerings can range from covering specialized niche tasks (e.g., network management, security management) to broader management packages that cover install, configuration, monitoring, and management tasks.
• Vendor-managed SaaS: Like other SaaS-based solutions (e.g., SFDC, etc.), these offerings allow you to get a cloud MDM environment up without the hassle of having to implement and manage the platform. However, because these offerings are built on a series of deployment decisions that the SaaS vendor has made, they may not afford you the flexibility to adapt easily to your needs. For example, the service may only work on a specific vendor platform, which may not line up with your company’s cloud standards. SaaS offerings also have restrictions on the type of access that you will be granted. You will also need to make sure the security and privacy controls are up to par and are compliant with your enterprise standards, regulatory, and data residency requirements. Most SaaS-based solutions are multitenant by design, which may raise security questions. Depending on usage style and volume, you will also want to check the performance and scalability factors, especially if you have heavy real-time transactional traffic. It’s essential to ensure that the SaaS MDM offering that you are evaluating meets your requirements. It’s also equally important to plan for the complete cloud migration/adoption lifecycle as part of your evaluation. Even with a fully managed SaaS offering, it can be a significant effort to implement an enterprise MDM SaaS solution. Solution configuration, testing, data migration, and management of on-prem integration are important tasks and are not included in typical MDM SaaS offerings.

Cloud-Native Software Features

• Cloud-native MDM: These are newer MDM platforms that follow a microservices container-based design. These solutions are great for companies comfortable managing cloud technologies and have the applications and usages that will leverage the associated microservice design. These products can offer faster and more scalable deployments that leverage modern cloud dynamic scaling options. It is also important to note that because these products and platforms are new, they may not have the full breadth and reliability of traditional MDM implementations. As a result, application capabilities and architectural reliability may be missing or not fully implemented yet. Cloud-native MDM architectures are the future, but it’s essential to assess the functionality delivered and your ability to implement this type of architecture. Your organization may not have the sophistication to maintain and leverage the microservices design of these newer cloud-native MDM solutions.
• Legacy MDM: These are older MDM products (e.g., IBM MDM, Informatica MDM, Semarchy xDM, etc.) that use technology stacks and design principles that are monolithic. The vendors may have ported these MDM applications to run on cloud platforms, but they are predominantly monolithic in nature. These solutions have traditionally proven MDM feature capabilities and reliability. Typically, they are great fits for complex business requirements and are proven in many ways, including performance, scalability, security, etc. Nevertheless, you will need to make sure the vendors have development and support plans for them on the cloud, and they will not be out of support anytime soon. Hopefully, the capabilities are rich enough that you won’t have to make major enhancements post-implementation, as their deployment typically will be more involved and longer.
• Hybrid cloud/legacy MDM architectures: These are older MDM products that have been “re-platformed” to a newer container-based architecture but have not been rebuilt to fully deliver a microservice-based solution. These solutions will take advantage of certain cloud capabilities (e.g., container auto scaling) but have architectural limitations that prevent them from leveraging other cloud capabilities (e.g., multi-availability zone implementations). It is also important to note that some of the older MDM products on the market may be providing a “mixed” solution architecture where portions of the MDM platform are fully microservice-container-enabled, but other portions of the solution are containerized implementations of older legacy architectures.
  
  Those of you who have an existing on-prem MDM running and are considering joining the journey to cloud will also need to decide if you are migrating or re-platforming:
• Migrating an existing on-prem MDM implementation to the cloud: This option involves moving your current on-prem MDM instance(s) to a new cloud environment that is based on the same underlying MDM software. This option preserves your sunk MDM investment while gaining some infrastructure efficiencies via vendor-provided cloud services. You should check with your existing MDM vendor to make sure this is supported. If you are familiar with cloud technology (e.g., management of the platform, network, security) and have an existing cloud presence, implementation requires minimum hassle, and it allows you to leverage the advantages of cloud. Companies with fast-growing volumes in their MDM usage will find the agility and lower platform cost on the cloud very appealing. If you are new to cloud deployments, it is advisable to work with an IT service company that has done this before.
• Re-platform to a new MDM on cloud: This option consists of retiring your current MDM implementation and picking a new solution, which requires a complete rebuild of your MDM solution on the new MDM cloud platform. This option is appealing when your existing MDM solution is not providing you the capabilities you need or your MDM vendor does not support running their solution on popular cloud platforms. You will have a similar decision process as a first-timer and need to decide between the various SaaS and/or cloud-native options for running MDM on the cloud. This is the more disruptive option, so you should carefully evaluate as there is really not much ROI to be made by changing MDM solutions. You might end up spending a lot of money, encountering new issues, incurring a long migration phase, and having limited enhanced capabilities over your old MDM.
• Finally, if you find the current MDM on cloud solutions offered by vendors are not viable for your requirements and decide to build one, then develop it for cloud. It would be best if you built it using container-based microservices architecture, preferably on a graph database. While this approach offers the lure of ultimate alignment to requirements, it is also not for the lighthearted. Development of next-generation MDM solutions requires extensive infrastructure and cloud skills over and above MDM domain and software development skills.
• Getting onboard the cloud journey is worth considering due to the proven benefits of cloud platforms versus on-prem implementations. Of course, this is not for everybody, as there are usage styles, security, and regulatory concerns for some. The cloud choices are many, complex, and not always obvious. It is advisable to consult an IT Data Management services company that specializes in MDM, has cloud experience, and can provide critical Data Management and cloud services. Look for a company that has done this many times before and can offer MDM assessment packages such as:
• MDM on cloud
• MDM move to cloud
• MDM upgrade
  MDM and cloud experts will typically have cloud adoption frameworks for MDM to assist you in picking the right cloud technology and cloud migration approaches.
• Additionally, if you are not implementing a fully managed SaaS offering, engaging an MDM services vendor to provide cloud, software, and operational support services for your MDM cloud platform can ease your transition. This way, you lower your risk for success and have access to a pool of resources with MDM and cloud expertise. Look for MDM services providers who provide MDM Center-of-Excellence (CoE) style offerings that can offer you expertise in MDM implementation, cloud migration, and cloud operations services.
• Starting your MDM cloud journey is not hard, but there are many options to consider that can significantly impact your cost and time to value. Engaging an IT Data Management company that offers MDM on cloud expertise would be a worthwhile investment to help you successfully start this journey.