Local Performance vs. End-to-End Optimization
Establish shared end-to-end service telemetry as the governing reference so local domain optimisation cannot shift congestion or latency undetected across the service path.
CyberTRIZ analysis · Telecommunications contradiction NC019 · one of 8,235 worked contradictions published by CyberTRIZ.AI
Regulations
Business Context
Individual telecommunications domains are frequently managed according to their own performance objectives. Radio teams optimize radio utilization, transport teams optimize routing and capacity, cloud teams optimize computing efficiency, and core teams optimize their respective functions. These local improvements can conflict with the performance of the complete service path. A decision that improves one subsystem may shift congestion, latency, cost, or failure exposure elsewhere.
Telecommunications TRIZ Resolution
Optimization should use end-to-end service objectives as the governing reference while allowing local domains to optimize within those boundaries. Shared telemetry, service models, dependency maps, and cross-domain control mechanisms can identify when a local improvement produces deterioration elsewhere. Domain optimization remains valuable but becomes subordinate to overall service performance.
Applicable TRIZ Principles
Principle 5 – Merging combines information from multiple network domains into a common performance view.
Principle 13 – The Other Way Round evaluates optimization from the service perspective rather than exclusively from individual infrastructure domains.
Principle 23 – Feedback uses end-to-end measurements to correct local actions that degrade overall performance.
Expected Outcome
Better end-to-end service performance
Fewer cross-domain optimization conflicts
Improved resource coordination
More accurate engineering decisions
Decision Indicators
Early indicators include:
Domain KPIs improve while customer performance remains unchanged or deteriorates.
Congestion repeatedly moves between network layers.
Engineering teams optimize against different objectives.
Root-cause analysis reveals problems created by otherwise successful local changes.
No common end-to-end performance model exists.