2021. okt. 22.6 perc olvasás

EA Repository : Using discovery to save time and increase quality

The post-pandemic recovery is underway. According to McKinsey, 75% of North American and European executives expect increased investment in new technologies through 2024. As organizations invest more in digitization, decision-making must be based on quality data and information that is readily accessible. Enterprise data quality is rooted in a strong and well-maintained Enterprise Architecture (EA) repository that uses discovery to save time and increase quality.

What is an EA repository and what is the benefit?

An EA repository is a collection of artifacts that describes an organization's current and target IT landscape. It is intended to provide a centralized place for the storage and retrieval of architecture artifacts. An EA repository should have the key logical components to the architecture metamodel, domains, principles, capabilities, governance documents, and reference library.

Organizations benefit from having a clear and documented approach or framework in which to conduct enterprise architecture projects including artifact templates, capability maps, and more. It stores the “as-is” architecture so gap analyses can be done to build “to-be” architectures and help identify opportunities to shorten time-to-market, reduce costs, and identify operational and technology improvements.

EA repository management

With all of these components, it’s no small undertaking to get the right information into the EA repository – as well as keeping it updated (critical for EA success). Enterprise architects are drowning trying to keep up with data accuracy, unable to reach the finish line as companies are dynamic and ever-evolving, and thus the finish line keeps moving further. This frustration is what architects mean when they talk about “death-by-repository.” In this scenario, EAs spend all their time trying to keep an "as-is" architecture up-to-date, and thus, do not have the bandwidth to practice "to-be" architecture that is focused on business outcomes.

As architects build and maintain an EA repository they may be doing this manually by importing spreadsheets from Excel, CMDB, etc. Of course, manual takes a lot of time and is prone to error. While it’s possible to begin building a common repository from scratch, this can become time-consuming, and EAs run the risk of falling behind. Further to this point, the design of the repository needs specific consideration, given many of the people using the repository won't have the everyday knowledge of where and how to find information. With proper tooling, however, there are easier ways EAs can begin to populate and design the EA repository.

With discovery and connectors that reach across technologies creating a common language, death-by-repository fears are eliminated. At its very core, the purpose of an EA repository is to drive connectedness to enable common insights, and overviews of relationships and interdependencies. Reducing the friction of keeping an EA repository updated benefits the entire company not only with more accurate data for stakeholder decision-making, but it also allows EA to do other more value-oriented projects that support business outcomes. Further, consider that on average, developers spend only 5% of their time writing new code, 20% modifying the legacy code and up to 60% understanding the existing code. Reducing time and manual work will free up time for developers and architects to focus on “to-be” future scenarios.

Some examples of different types of discovery

Enterprise Architecture deals with many concepts including applications, technologies, processes, and data. To facilitate the population of an EA repository, there are four types of discovery to build, manage, and optimize information that can save considerable time and costs – and increase the quality of data and decision-making.

Application discovery is a process through which applications installed and used throughout an enterprise are identified and collected. It provides visibility into server activity, collecting and presenting configuration, usage, and behavior data from servers allowing one to get a total picture of server workload. Application discovery can reduce risk during code modernization by helping to understand the structure and interdependencies of legacy applications. It creates an inventory of all technical assets, which can help improve maintenance and operations, and delivers a strategy for future-state IT and modernization. Application discovery helps enterprise architects understand and manage metadata associated with each application and in turn is used to build an EA repository.
Cloud discovery automatically discovers all cloud instances spanning applications, databases, and related services running at a given point in time. As organizations continue to increase usage of the cloud, it is important to understand the enterprise’s cloud footprint to allow security and auditing functions the ability to know where and how to manage rogue and unmanaged deployments and identify shadow IT to reduce risk exposure. Cloud discovery plays an important role in helping enterprises that are struggling to manage their multi- and hybrid cloud instances by enabling enhanced visibility, improved optimization, and reduced risk to maximize return on cloud investments. Cloud discovery helps enterprise architects and technology innovation leaders understand their cloud footprint and where/how to manage potential gaps.
Process mining transforms data into event logs and then creates visualizations of the end-to-end process, along with insightful analyses, allowing you to compare what was designed with the actual process. It uses event logs to show what people, machines, and organizations are really doing and provides insights on how to improve and address performance, bottlenecks, and compliance breakdowns. Process mining helps enterprise architects and technology innovation leaders understand operations and performance in-order-to optimize processes.
Data discovery, in the context of IT, is an exercise that enables an enterprise to find its data stored in databases, build a data catalog and business glossary, and model data architecture. Data discovery is a process for collecting data from various sources by detecting patterns and outliers enabling the consolidation of business information. Based on this discovery, information architects can build a data dictionary, create a business glossary, and design data models. They can also perform data lineage analyses and ensure compliance to data regulations. Data discovery helps enterprise architects and technology innovation leaders visually understand where data sits and is leveraged within the organization.

Repository structure: APIs and Connectors used by EA tools

There are two ways to connect access points and thus share data, through: 1) Application Programming Interfaces (APIs); and 2) Connectors.

APIs are a set of functions and procedures that allow for the creation of applications that access data and features of other applications, services, or operating systems. It is the code that governs the access point(s) for the server. This is especially useful as the API allows one the ability to continuously query an application, get that data, and view or use it in another application.
Connectors are the pieces that connect to the API and pass that data to the next message processor as a data stream. A Connector is for an Application what a VPN is for a Network. They should only exist for proprietary applications that cannot expose services as APIs themselves.

An EA tool with a dynamic connected repository provides the ability to browse and drill down on relevant details to enable architects and developers to conduct impact analyses and plan migration projects. Tools that use APIs and Connectors to conduct discovery are especially useful as they allow enterprise architects the ability to start providing value on Day One.

Third-party integrations or connections with ServiceNow, CMDBs, Flexera, SAP, Salesforce, Minit, structured and unstructured databases such as MongoDB, and many other technologies help EAs build a repository instantaneously and maintain it with quality information.

When an EA tool uses APIs and Connectors to connect and build upon data, the possibilities are almost endless.

Discovery saves time and increases quality

Discovery is the population and maintenance of applications, processes, and data, so companies can manage and optimize their EA repository and feel confident about its accuracy and quality. It uses open APIs, integrations, and connectors that fetch the right data. Discovery can also be associated with artificial intelligence to detect patterns and perform advanced analyses.

The benefits are real:

It quickly populates the EA repository by connecting to operational data.
It ensures quick time-to-value, increases data accuracy, and can be relied upon for decision-making.
Reports and analytics can instantly be created and immediately leveraged.
Architects can use these insights to drive business intelligence and design “to be” states - and not have to spend time on "as is" current state, as it’s already done.
EA's perceived value inherently moves from noisy to influential.

To summarize, discovery not only speeds up the population of the EA repository, it frees up enterprise architects to focus on business outcomes that can support agile delivery and digitization. And ultimately, it ensures the EA repository is a single-source-of-truth for IT standards that can be relied upon by users and stakeholders to make smarter, faster decisions.

For more information, view MEGA’s eBook detailing the ROI of Business-Outcome-Driven Enterprise Architecture.