Capacity & Bottleneck Propagation
How growth or pressure in one layer creates saturation in another.
Identify services that will exhaust a resource within N days at current growth rate; include resource type, projected exhaustion date, and growth model.
2. How much headroom do we have before the next traffic peak?Determine remaining headroom before the next traffic peak given current capacity, replica counts, autoscaler limits, database max connections, cache capacity, and downstream rate limits.
3. What is the resource cost of adding N users across all layers?Map the resource cost of adding N users or N requests/sec across application, database, cache, message bus, and external API layers.
4. Are any resource types growing at a different rate than the others?Detect imbalanced scaling where one resource type (memory, connections, queue lag, IOPS) is growing faster than others; identify the multi-system bottleneck this implies.
5. What would consolidating these workloads do to utilization and shared dependencies?Model utilization impact of workload consolidation including anti-affinity constraints, noisy-neighbor risk, and shared dependency implications.