https://store-images.s-microsoft.com/image/apps.55310.7c90bc84-ca6c-4f0f-9bee-31f2cf60222a.5bca64ae-a613-4d62-ac42-580e21e01225.2a7bfc93-86bd-4ef9-a5f5-a7dd62a1bbca

Stibo Systems Data as a Service - DaaS

Stibo Systems A/S

Stibo Systems Data as a Service - DaaS

Stibo Systems A/S

Drive significant competitive advantage by making master data easily accessible

Stibo Systems Data as a Service (DaaS)

Drive significant competitive advantage by making master data easily accessible for operational, analytical and machine-learning decision-making processes.

Master data management (MDM) can secure a single source of truth of your data, enabling access to consistent and trusted key information across departments. In this way, MDM empowers retailers, distributors, manufacturers and financial institutions to enhance collaboration, operational efficiency, customer experiences, data analytics and many more operations that require trusted data. Governed master data can be delivered from the MDM to downstream systems, processes or parties who want to use that authoritative set of data.


Scalability challenges
In a traditional setup, it is common practice to build and maintain separate data stores with APIs between the MDM and the target systems to handle use cases that consume master data at very large scale, e.g., websites or point-of-sale systems. This provides a separation of concerns between complex master data processes and high-volume data consumption so that one activity doesn’t disturb the other. However, this entails a number of operability problems:

  • Building and maintaining a separate data store with associated API can be a cumbersome process.
  • Data stores may easily become inconsistent.
  • Allowing changes to master data to get through to consuming systems can take a lot of time since those changes may also necessitate development on the data stores.


The DaaS solution
Stibo Systems Multidomain MDM features a Data as a Service (DaaS) extension that can solve for the scalability issues. It’s a cloud-based data distribution service focused on the delivery of data to high-volume data-consuming applications. Through its configurable API and serverless architecture, the DaaS extension delivers a near real-time version of master data. Users don’t need to build and maintain numerous API services but simply re-configure the API. As a single, centralized master data service, the DaaS extension removes the need for creating multiple copies of data for each application in order to deliver data at scale to applications.

Always-on data delivery
The “as a service” means that you are not required to build and maintain a database with a copy of the data nor to build an API to deliver the data. The result is that you save time for maintenance and don’t need to worry about duplicate databases becoming inconsistent. Instead, you can just let your applications look up data synchronously or receive events on data asynchronously.

Service configuration
DaaS enables multiple services to be configured for the same master data, each with their own API, API keys and associated master data subset. Each API can be configured for maximum ease of consumption:

  • Easy to consume the data without having to understand the underlying data model.
  • No need to have the same data model in your application as in your MDM.
  • Fast and easy to adapt APIs to changes in your master data model.
  • You can change names of the objects and limit the scope of data to avoid exposing your entire data source to the application.

By making data available through an API service layer and serving data at scale on-demand for internal as well as external systems, the DaaS extension is a perfect example of the data-at-the-edge philosophy.


Benefits of MDM with DaaS

  • Always-on access. Data is available to consume through an always-on service where the recipient is in control.
  • High volume scalability. Data consumption may scale to thousands of parallel requests.
  • Near real-time. Data and updates to data can be obtained in near real-time – no replication or download of a copy needed.
  • Standards-based. Multiple systems can consume from the service through standards-based cloud-native technology.
  • Easily configurable APIs. APIs are separate from the MDM data model in order to make it easy to consume data.
  • Improved performance. High volume requests place less strain on internal systems and data can be served with minimal latency.
https://store-images.s-microsoft.com/image/apps.2064.7c90bc84-ca6c-4f0f-9bee-31f2cf60222a.5bca64ae-a613-4d62-ac42-580e21e01225.76dbbc8c-f140-4841-8a8a-5924002d1ae1
https://store-images.s-microsoft.com/image/apps.2064.7c90bc84-ca6c-4f0f-9bee-31f2cf60222a.5bca64ae-a613-4d62-ac42-580e21e01225.76dbbc8c-f140-4841-8a8a-5924002d1ae1
https://store-images.s-microsoft.com/image/apps.50417.7c90bc84-ca6c-4f0f-9bee-31f2cf60222a.5bca64ae-a613-4d62-ac42-580e21e01225.540132df-4548-487e-9bd2-a9bb5901257d