Why your IT Inventory is a mess

Your IT Inventory a hot mess. Here is why

In the 90s, along with ITIL, came a concept of CMDB, that kinda solved the infrastructure inventory problem… but CMDB was destined for failure from the start, mostly because the only mechanism to ensure the data quality were humans manually ingesting data.

How do you currently handle IT Asset Inventory? Is it CMDB? Ask yourself these questions:

• Do your Developers have a way to directly interact with CMDB data?
• Is Data Modification a self-service, so that each team can be responsible for their data quality?
• Can they modify the data during deployments in an automated and autonomous way (self-service, instead the open-a-ticket-to-infra-team)?
• Can they build their own APIs using the Inventory Data?
• Can they build dashboards and explore the Inventory Data?

If the answer to any of these questions is NO, your current solution (CMDB, or whatever you might have) just wont do. You need something else…

Requirements for building the “something else”

What are the Requirements? We need 360 vision of our application. We must guarantee automated data injection to guarantee data quality, and self service to make the Asset Owners responsible for the data that needs to be managed manually. Last, but not the least, we need DATA.

Data we can inject automatically:

• Software Stack, used languages, versions, libraries
• Application sub-structure, with microservices or “legacy” sub-components
• Infrastructure
• Licenses
• Contracts with all the discounts and support costs
• Automatically discovered dependencies with other Apps / Shared Infrastructure

Data we need Asset Owners to maintain manually updated:

• Description of what the Application does
• Relevant stakeholders
• Web URL, and other Access information
• Legal, Business and Data Protection requirements
• Owners INTENT: Do you want to Modernize, Migrate or Decommission this application? When?

Solution: Demystify the “Something Else”

There are 4 “parts” of what you actually need to build:

1. Deploy the Data System/Platform
2. Get your Data into the Platform
3. Get the Funding
4. Accelerate using AI

1 - Data System. Relatively easy

You need the right internal data platform architecture, that allows you using the best “features” of a Data Lake, Data Fabric and a Data Mesh. It can be done with a team of good Data and Cloud Architects and Engineers.

The architecture will highly depend on your organization. An example of a good approach could be:

• Store raw data in a data lake: This gives you a centralized repository for all your data, regardless of its format or structure.
• Use a data fabric to connect to and integrate data from multiple sources: This allows you to connect to your data lake, as well as other data sources such as relational databases, cloud applications, etc.
• Adopt a data mesh for distributed data governance: This gives domain teams ownership of their data, which can improve data quality and agility.

Once your data solution is in place, you can create the Data Products, such as “Configuration Management API”, or “360 Infra Exploration Tool”, or “Cost Explorer”.

2 - Get the data. Challenging

Definitely not the 100% of the of all our IT system. Find a “sweet spot” between PERFECT and NOTHING.

3 - Get your organization buy-in. Extremely difficult

You will go directly against the “this is how we’ve always done things” group, so… prepare. Use this blog post to build your case.

4 - Don’d forget the AI

Combining the right functionalities of Data Lake, Data Fabric and Data Mesh, with AI capabilities, can enhance your IT asset inventory management and provide valuable insights for optimizing your IT infrastructure. Here are some compelling use cases for AI in this architecture:

1. Automated Asset Discovery and Classification

AI algorithms can automatically scan the data lake and data fabric to identify and classify IT assets, including hardware, software, and network devices. This can help reduce the time and effort required for manual asset discovery, ensuring that your IT inventory is always up-to-date and accurate.

2. Proactive Risk Detection and Mitigation

AI can analyze asset data to identify potential risks, such as outdated software, missing patches, or vulnerabilities. This early detection can enable proactive remediation efforts, preventing security breaches and downtime.

3. Predictive Capacity Planning

AI can analyze historical trends and usage patterns to predict future IT resource needs, such as storage, compute, and network capacity. This foresight can help you optimize your IT infrastructure investments and avoid potential bottlenecks.

4. Enhanced Compliance and Auditing

AI can help you ensure compliance with industry regulations and internal policies by analyzing asset data for compliance gaps. This can help you maintain a secure and compliant IT environment.

5. Cost Optimization and Procurement

AI can analyze asset data to identify underutilized or obsolete equipment, helping you optimize your procurement decisions and reduce IT costs. This can free up resources for other strategic IT initiatives.

6. Personalized IT Support

AI can analyze user behavior data to provide personalized IT support recommendations, such as troubleshooting tips or suggested software upgrades. This can enhance user experience and improve IT service delivery.

7. Predictive Maintenance and Asset Replacement

AI can analyze asset data to predict potential hardware failures, enabling proactive maintenance and replacement schedules. This can minimize downtime and reduce maintenance costs.

8. Real-Time Asset Monitoring and Alerting

AI can continuously monitor asset data to identify anomalies or potential issues, triggering alerts to IT teams for timely intervention. This can help prevent downtime and ensure the reliability of your IT infrastructure.

By leveraging AI within your data lake and data fabric architecture for IT asset inventory management, you can gain deeper insights, make informed decisions, and optimize your IT operations for greater efficiency, security, and compliance.

Concepts

• CMDB (Configuration Management Data Base) is a repository of information about the IT infrastructure of an organization. It is used to store information about all of the IT assets in the organization, including hardware, software, network devices, and even personnel. The CMDB is a critical tool for IT operations, as it provides a single source of truth for all IT information.
• Data Lake is a centralized data repository, that stores data in its raw, native format.
• Data Fabric is a software architecture that connects and manages data from disparate sources. It’s a unified view of data across the enterprise, making it easier to find, access, and use. Check out more info about the concept and value of Data Fabric.
• Data Mesh is a decentralized approach to data management. Data is owned and managed by domain-specific teams, rather than a centralized data team (like Data Lake). It’s Self-Service and Domain Oriented, while the governance model is Federated. It promises higher data quality and lower costs, but the implementation can be complex.

Summary Video

• ← Previous
Next →