This blog post explains the aspects if interface/integration management, what it is about, why it is needed and which value it provides. Finally it also highlights what is the alternative to each aspect and what consequences result from an company´s behavior.
Inventory (Transparency)
- WHAT: An automated inventory/catalog/list of your interfaces which is always up to date
- WHY: Be ready for any analysis on interfaces in your Integration layer
- VALUE: Efficiency for the IT team: be able to plan and perform changes faster
- ALTERNATIVE: Do it manually and accept this not being up to date or only when this information is requested (acting reactively)
Documentation
- WHAT: An automated technical Documentation of your interfaces which is always up to date
- WHY: Knowledge transfer for new colleagues, fulfill audit requirements, quick check on integrations
- VALUE: Efficiency for the IT team: no more manually created technical documentations
- ALTERNATIVE: Do it manually and accept this not being up to date
Reporting (Traffic)
- WHAT: An automated report of the messaging traffic (number of messages, e.g. for the last month or week)
- WHY: Know the intensity of your integrations (how many messages are we exchanging?) helps to challenge some integrations
- VALUE: Manage complexity: Know what can be decommissioned
- ALTERNATIVE: Collect this information manually or do not care
Reporting (Performance)
- WHAT: An automated report of the messaging performance (duration of integrations, e.g. for the last month or week)
- WHY: Know the speed of integrations (how long does the processing take place including transformations, transport/connectivity and orchestration) to improve the processing if required
- VALUE: Robustness: Identify slow integrations, bottlenecks to avoid downtimes
- ALTERNATIVE: Collect this information manually or do not care
Reporting (Volume)
- WHAT: An automated report of the messaging volume (data transferred, e.g. for the last month or week)
- WHY: Know how much data is flowing between applications/systems/partners (in Bytes)
- VALUE: Allow automated cost allocation to divisions/brands/domains of the organization, challenge the data volume exchanged across systems
- ALTERNATIVE: Collect this information manually or do not care
Checks (Quality)
- WHAT: Perform Design Time checks on the integration content, such as naming conventions or typical integration design errors
- WHY: Ensure a high quality of the integration content from the very beginning and establish quality gates
- VALUE: Automated checks during the development of interfaces increase the quality and lower downtimes
- ALTERNATIVE: Perform checks according to a checklist manually (4-eye-principle) or trust all developers to deliver highest quality by default
Checks (Landscape)
- WHAT: Perform a comparison of integration content versions across several middleware systems (e.g. development, quality, validation, production)
- WHY: Make sure to have integration content in sync and lower the risk of missing out a transport or deployment across your integration layer landscape
- VALUE: Automated checks across the landscape increase the quality and lower downtimes
- ALTERNATIVE: Perform the checks eventually manually (weekly housekeeping) or do not care and trust in your developers and transport management
Checks (Certificates)
- WHAT: Find certificates which are about to expire in the coming XX days and get notified (alerting)
- WHY: Make sure you replace certificates in time before they expire and your integrations stop working
- VALUE: Automated checks on your certificates help to avoid downtimes and ensure business continuity
- ALTERNATIVE: Perform the checks manually (weekly housekeeping) or wait for the interfaces to stop and replace the certificate after a (hopefully short) downtime
Checks (Errors)
- WHAT: Find blocked messages (in error) or channels failing and get notified (alerting)
- WHY: Be informed when an interface stops working
- VALUE: React quickly to reduce the downtime of an interface to ensure business continuity
- ALTERNATIVE: Wait for the business to complain and create a ticket/incident
Checks (Anomalies)
- WHAT: Find unusual traffic (too many API calls/no messages at all) and get notified (alerting)
- WHY: Be informed when an interface behaves strangely which is an indicator of errornous situations
- VALUE: Anticipate error situations and prevent downtimes of an interface to ensure business continuity
- ALTERNATIVE: Wait for the business to complain and create a ticket/incident
Please consider that there are also benefits for an integration team like being perceived more professionally by other teams as well as an increasing self-perception (being enabled through quality improving tools and automations to free their time for projects and initiatives).
