IT Service Management: Does the Provance ITSM CMDB also include Cloud Services information?
The simple answer is yes, of course. A CMDB is a fundamental component in the ITIL standard. The tricky part is what is a CMDB? A configuration management database (CMDB) is a repository that acts as a data warehouse for IT organizations. It contains a collection of IT assets/resources that are commonly referred to as configuration items (CI), as well as descriptive relationships between the CIs. Most commonly, we think of software, hardware, network, and storage CI types (sometimes referred to as discovery CIs) as being stored and tracked in a CMDB. As enterprises mature, they want to also track business CIs in their CMDB such as: organizations, people (users and contacts), services, products, and 3rd party entities such as vendors and partners. Now because of the cloud, enterprises will need to track cloud applications and resources and therefore Provance ITSM has introduced Cloud CIs into the CMDB.
New tools, such as Azure security & operations management (formerly known as Operations Management Suite) are available to manage both the cloud and on-premise CIs. This provides us with new opportunities to increase both the operational and planning aspects associated with the Provance CMDB. The complex nature of the cloud increases not only the number and types of CIs but also the dynamic nature of the data.
In the ITIL framework, the CMDB is part of infrastructure operations and support and it represents the authorized and actual configuration of the significant components of the IT environment. It tracks CIs at any point in their existence for analysis and support of the five major processes:
- Service Definition (i.e. Availability Management)
- Service Transition (i.e. Change Management)
- Service Operation (i.e. Incident, Event and Problem Management)
- Service Strategy (i.e. Financial Management)
- Continual Service Improvement (i.e. Service Review)
So we have a repository which has business CIs and their supporting CIs including discovery CIs. CMDB implementations often involve federation, the inclusion of data into the CMDB from other sources in such a way that the source of the data retains control of that data. Federation is usually distinguished from ETL (extract, transform, and load) solutions in which data is copied into the CMDB. ETL processing allows data from various sources to be standardized and merged. It can be summarized and it may be processed into facts, dimensions and measures (data warehouse concepts) for trending and analytics. The Provance CMDB is a primarily an ETL solution, we use connectors to copy and process data from operational databases like System Center Configuration Manager or Active Directory. However, with dynamic data sources such as the cloud, the Provance CMDB has added the ability to access data in real-time as well as use ETL processing to link business and discovery data. The Provance CMDB organizes that data and provides both a snapshot (when the connector ran) and a change history over time for selected CIs as well as dynamic real-time access via hyperlinks in the user interface in a federated view.
Azure security & operations management has the ability to monitor new types of CIs, such as subscriptions, resources and cloud-based software (software-as-a-service, platform-as-a-service, or infrastructure-as-a-service). We can import this data and tie it to the business information we have in Provance CMDB. For the Service Catalog—the services that we provide—we can now see, discover and document CIs that are part of our service offering, even if they’re in the cloud; assuming you have Azure security & operations management. Of course, you may not want to bring all the data from Azure security & operations management over into the CMDB, but you may want to reference it and the Provance CMDB allows you to do so. Additionally, the Service Maps functionality in Azure security & operations management, discovers the connections between devices via the device hardware services. For example, you can see that a particular server is running SQL server and there are also several other devices that are utilizing a SQL server connection to that device. Unlike Azure security & operations management, the Provance CMDB has the full context associated with the services: what the connection is, how important it is to the organization and what the service is for. All of this type of information can be documented with the Provance CMDB.
CMDBs typically have one of two data models either semantic or relational. A semantic data model allows anything to be related to anything else by creating a relationship descriptor and a means to associate individual records by abstraction. It’s highly flexible and customizable but is very complex to build and maintain in terms of code and infrastructure and it requires a significant understanding in order to use in reporting and queries. Relational data models are simpler in that related records are linked together via a key. Challenges lie in extending the data model and creating “new” or not in out of the box relationships. However, in terms of capacity and performance it can leverage the power of database platforms like SQL. The Provance data model is relational, so we allow SQL to do what it does best, store and retrieve data efficiently and process massive amounts of data effectively. One of the challenges of harvesting and maintaining discovery data is the volume and volatility, simplicity equals performance.
We believe the key success factors in implementing a CMDB where discovery data plays a large role as it does in ITSM, lie in five areas:
- ability to automatically discover information about the CIs (auto-discovery) and track changes as they happen
- ability to support batch (connectors) and real time (events) updates to CIs
- ability to keep track of the state of CIs and their relationships as they exist at specific points in time
- ability to facilitate the creation of facts – measures and dimensions over the history of the CI
- ability to reduce the operational impact of the volatility and volume of the supporting CIs versus the highly structured business CIs
Based upon this criteria, Provance has built its CMDB across two databases. Using Dynamics 365 (Primary Store) to house business CIs while utilizing a second SQL database, the PMDB (Backing Store), to hold transient data and supporting CIs. This allows us to keep the primary data elements (business CIs) close to our user interface and leverage Microsoft Dynamics 365 strengths and integrations while supporting secondary request data elements like discovery items or more voluminous and less referenced items physically separate to isolate operational impacts and processing noise from the user experience. Additional discovery data like the Cloud CIs and its dynamic references are also held in the Primary Store. The PMDB can be located either on premise or in the cloud depending on your configuration and security requirements.
The Provance CMDB is a single logical repository structured around performance and flexibility, it is a hybrid solution addressing some of the fundamental pain points and design flaws in implementing a CMDB. So yes, we have a CMDB and it’s built to perform and to adapt to changing requirements.