Latest Evidence Run 🔵 Measured
2026-03-09 23:32 UTC
Timestamp of the latest evidence bundle used for this report.
Uptimer Capability Evidence
Stats for Nerds
Enterprise evidence pack combining measured benchmark runs, modeled capacity projections, and live anonymized runtime telemetry.
Evidence Run: 20260309-233210
Recorded: 2026-03-09 23:32:47 UTC
Latest local evidence bundle
Loaded from artifacts/test-reports/performance/20260309-233210
Section 1: Evidence Classification
🔵 Measured
Created directly from benchmark artifacts in artifacts/test-reports/performance/<runId>, including
execution-throughput, scheduler-scale, and execution-reliability scenario outputs.
Values are parsed from raw JSON and surfaced without projection math.
🟡 Modeled
Built from measured baseline plan throughput (plan-capacity-projection) and then scaled using explicit controls:
selected plan capacity, active-tenant count, and deployment multiplier (cloud = 1.0x, self-hosted = selected factor).
Daily and weekly values are deterministic transforms from checks/sec.
🟢 Live Computed from current production database counters over rolling windows (for example checks completed in the last 24 hours), then normalized into rates (checks/sec) and load ratios against modeled capacity. These are anonymized aggregate telemetry snapshots, not synthetic test output.
🟣 Theoretical Envelope
Derived from the max-usage-envelope scenario by applying declared workload bounds
(min/max concurrent checks and fastest/slowest intervals) to produce upper-limit demand and capacity envelope estimates.
Use this as a planning boundary, not an observed runtime guarantee.
Section 2: Platform Status Overview
Result 🔵 Measured
PASS
Evidence integrity verified; reliability within control limits.
Observed Sustained Throughput 🔵 Measured
76.3 checks/sec
Sustained execution rate from the measured benchmark workload.
Reliability Control Status 🔵 Measured
Within target threshold
CV target: 0.350, observed CV: 0.288.
Scheduler Saturation 🟡 Modeled
0.0%
Derived from observed queue size relative to expected queue burst target.
Queue Fairness 🔵 Measured
Verified
No tenant starvation means no tenant queue was continuously deprioritized while others progressed.
System State 🟢 Live
Stable under multi-tenant load
Posture derived from live throughput and absence of queue pressure indicators.
Conclusion: Uptimer demonstrates predictable scaling behavior under bounded concurrency with significant headroom relative to current production usage.
Hardware Profile of Evidence Run
🔵CPU Class: 12 logical cores
Memory Class: 62.7 GiB
Database Engine (Benchmark Profile): SQLite
Storage: Local fixed volume
OS Family: Ubuntu 24.04.4 LTS
Runtime: .NET 10.0.3
Section 3: Measured Performance Validation
Observed Execution Throughput 🔵
76.3 checks/sec
Measured sustained throughput during controlled 400-monitor parallel test (parallelism: 6). Represents stable steady-state processing rate without queue backlog.
Reliability P95 Throughput i 🔵
97.7 checks/sec
95th percentile throughput across 3 independent benchmark rounds. Coefficient of Variation: 0.288. Lower CV indicates better repeatability.
Scheduler Fairness i 🔵
Verified
40 tenants simulated. Queue burst target achieved at 100.0% expected fill. Tenant starvation means one tenant repeatedly waits while others continue to run.
Section 4: Modeled Multi-Tenant Capacity Projection
ModeledThese figures are projections derived from measured baseline throughput and scaled concurrency assumptions. They represent expected sustainable throughput under equivalent hardware scaling.
Selected Tenant Scale (Active Tenants = 300)
Per-Tenant Modeled Capacity 🟡
66.7 checks/sec
Per-Tenant Modeled Capacity (BUSINESS plan).
Modeled Throughput at Selected Scale 🟡
20,000.1 checks/sec
Selected-tenant throughput using current controls and deployment mode.
Daily Projection (Selected Scale) 🟡
1,728,008,640 checks/day
Modeled checks/sec multiplied by 86,400 seconds/day.
Weekly Projection (Selected Scale) 🟡
12,096,060,480 checks/week
Modeled checks/day multiplied by seven days.
Safety Margin vs Measured Test (Selected Scale) 🟡
262.3x
Selected-scale modeled throughput divided by measured execution throughput.
Reference Scale (300 tenants, benchmark baseline)
Modeled Multi-Tenant Cloud Baseline 🟡
20,000 checks/sec
At-scale benchmark baseline throughput from plan evidence.
Daily (Reference Scale) 🟡
1,728,000,000 checks/day
Reference checks/sec multiplied by 86,400 seconds/day.
Weekly (Reference Scale) 🟡
12,096,000,000 checks/week
Reference checks/day multiplied by seven days.
Section 5: Workload Coverage Analysis
Theoretical EnvelopeThe selected workload Theoretical Envelope (100 monitors at 30s) requires 3.3 checks/sec. Current modeled capacity 🟡 Modeled provides 20,000.1 checks/sec. This configuration provides 6,000.0x excess capacity over this workload.
Requirement Gap: -19,996.8 checks/sec (required - capacity; negative means surplus capacity).
Operationally, this demand can be absorbed without measurable queue pressure when model assumptions hold.
Section 6: Scheduler Concurrency and Backpressure Controls
Scheduler fairness uses tenant round-robin batching so due checks are distributed across tenants before deep-draining any single tenant backlog.
Location routing is stress-aware: automatic placement picks the lowest-stress location within each tenant's plan-allowed slots, across built-in globals and configured nodes.
Current configured location slots: 8 (includes cloud built-ins plus configured node locations).
Execution control combines global/per-tenant/per-host limits with bounded queueing and per-monitor de-duplication to prevent slot starvation and duplicate in-flight runs.
Concurrent Check Slots i 🔵
30
Configured number of simultaneous checks in the scheduler benchmark.
Queue Burst Capacity 🔵
120
Expected aggregate burst target (`maxConcurrentChecks × 4`) used to validate burst absorption across queue partitions.
Observed Aggregate Queue Fill i 🔵
120
Observed aggregate queue count at benchmark time (100.0% of target).
Active Queue Partitions 🔵
3
Distinct queue partitions sampled during the scheduler benchmark (global locations plus node queues).
Queue Distribution Imbalance 🔵
0
Difference between busiest and lightest sampled queue partition during the burst window.
Max Observed Queue Delay 🟡
1.57s
Derived upper-bound queue wait using worst sampled partition depth and per-partition throughput estimate.
Average Queue Wait Time i 🟡
0.79s
Derived midpoint wait under even queue service, used as a practical waiting-time indicator.
Longest Queue Drain Time 🟡
1.57s
Derived drain time for worst partition burst at estimated per-partition throughput.
Section 7: Plan Throughput Visualization
Plan rows lead with service envelope and intended profile; throughput values below are modeled at-scale cloud baseline values from benchmark evidence.
| Plan | Max Monitors | Min Interval | Modeled At-Scale Throughput (Cloud Baseline, 300 tenants) | Intended Usage Profile |
|---|---|---|---|---|
| FREE | 50 | 120s | 125.0 checks/sec 🟡 | Low-frequency non-critical checks |
| SOLO | 100 | 60s | 500.0 checks/sec 🟡 | Small workloads with moderate cadence |
| TEAM | 250 | 30s | 2,500.0 checks/sec 🟡 | Shared team operations and service groups |
| BUSINESS | 1,000 | 15s | 20,000.0 checks/sec 🟡 | High-density multi-service production monitoring |
Section 8: Current Production Telemetry Snapshot
This is real runtime data, aggregated and anonymized. It is not synthetic test data.
Current Production Load 🟢
0.0848 checks/sec
Derived from rolling 24-hour completed checks (7,323 in last 24h).
Tested Sustained Load 🔵
76.3 checks/sec
Measured controlled benchmark throughput.
Modeled Multi-Tenant Cloud Baseline (300 tenants) 🟡
20,000 checks/sec
At-scale modeled throughput under benchmark tenant assumptions.
Load-to-Capacity Ratio 🟢
0.000424%
Current production throughput as a percentage of modeled capacity.
Live stats recorded at 2026-03-10 00:16:10 UTC
Risk and Constraints
- Modeled figures assume similar hardware, network conditions, and monitor mix as measured runs.
- Queue delay metrics in this report are derived estimates, not direct per-item queue latency traces.
- Hardware profile fields should be explicitly pinned for each evidence run when used in procurement or external audits.
- Live telemetry is aggregated and anonymized; it indicates load posture, not per-tenant behavior.
Raw Test Evidence (Verbatim)
Direct JSON outputs from the performance test pack.
plan-capacity-projection
{
"ActiveTenants": 300,
"HighestFinitePlan": "business",
"HighestFinitePlanMaxMonitors": 1000,
"HighestFinitePlanMinIntervalSeconds": 15,
"HighestFiniteChecksPerSecondPerTenant": "66.667",
"HighestFiniteChecksPerDayPerTenant": "5760000.000",
"HighestFiniteChecksPerWeekPerTenant": "40320000.000",
"HighestFiniteChecksPerSecondAtScale": "20000.000",
"HighestFiniteChecksPerDayAtScale": "1728000000.000",
"HighestFiniteChecksPerWeekAtScale": "12096000000.000",
"Projections": [
{
"PlanCode": "free",
"MaxMonitors": 50,
"MinIntervalSeconds": 120,
"ChecksPerSecondPerTenant": "0.417",
"ChecksPerDayPerTenant": "36000.000",
"ChecksPerWeekPerTenant": "252000.000",
"ProjectedChecksPerSecondAtScale": "125.000",
"ProjectedChecksPerDayAtScale": "10800000.000",
"ProjectedChecksPerWeekAtScale": "75600000.000"
},
{
"PlanCode": "solo",
"MaxMonitors": 100,
"MinIntervalSeconds": 60,
"ChecksPerSecondPerTenant": "1.667",
"ChecksPerDayPerTenant": "144000.000",
"ChecksPerWeekPerTenant": "1008000.000",
"ProjectedChecksPerSecondAtScale": "500.000",
"ProjectedChecksPerDayAtScale": "43200000.000",
"ProjectedChecksPerWeekAtScale": "302400000.000"
},
{
"PlanCode": "team",
"MaxMonitors": 250,
"MinIntervalSeconds": 30,
"ChecksPerSecondPerTenant": "8.333",
"ChecksPerDayPerTenant": "720000.000",
"ChecksPerWeekPerTenant": "5040000.000",
"ProjectedChecksPerSecondAtScale": "2500.000",
"ProjectedChecksPerDayAtScale": "216000000.000",
"ProjectedChecksPerWeekAtScale": "1512000000.000"
},
{
"PlanCode": "business",
"MaxMonitors": 1000,
"MinIntervalSeconds": 15,
"ChecksPerSecondPerTenant": "66.667",
"ChecksPerDayPerTenant": "5760000.000",
"ChecksPerWeekPerTenant": "40320000.000",
"ProjectedChecksPerSecondAtScale": "20000.000",
"ProjectedChecksPerDayAtScale": "1728000000.000",
"ProjectedChecksPerWeekAtScale": "12096000000.000"
},
{
"PlanCode": "enterprise",
"MaxMonitors": null,
"MinIntervalSeconds": 5,
"ChecksPerSecondPerTenant": "Infinity",
"ChecksPerDayPerTenant": "Infinity",
"ChecksPerWeekPerTenant": "Infinity",
"ProjectedChecksPerSecondAtScale": "Infinity",
"ProjectedChecksPerDayAtScale": "Infinity",
"ProjectedChecksPerWeekAtScale": "Infinity"
}
],
"RecordedAtUtc": "2026-03-09T23:32:36.8012132Z"
}execution-throughput
{
"MonitorCount": 400,
"Parallelism": 6,
"CompletedChecks": 400,
"MinimumChecksPerSecond": 8,
"ObservedChecksPerSecond": 76.258,
"ElapsedSeconds": 5.245,
"RecordedAtUtc": "2026-03-09T23:32:44.8749421Z"
}scheduler-scale
{
"ExpectedBatchSize": 120,
"ObservedQueueCount": 120,
"SampledTenantCount": 40,
"SampledUniqueMonitorCount": 120,
"SampledLocationCounts": {
"global-1": 40,
"global-2": 40,
"global-3": 40
},
"SampledLocationImbalance": 0,
"MaxConcurrentChecks": 30,
"MonitoringLocations": "global-1,global-2,global-3",
"RecordedAtUtc": "2026-03-09T23:32:47.7102908Z"
}execution-reliability
{
"Rounds": 3,
"MonitorCountPerRound": 300,
"Parallelism": 6,
"MinimumChecksPerSecond": 8,
"MaxAllowedCoefficientOfVariation": 0.35,
"AverageChecksPerSecond": 96.886,
"P50ChecksPerSecond": 97.668,
"P95ChecksPerSecond": 97.668,
"MinChecksPerSecond": 62.332,
"MaxChecksPerSecond": 130.658,
"ThroughputCoefficientOfVariation": 0.288,
"ThroughputSamples": [
62.332,
97.668,
130.658
],
"ElapsedSecondsSamples": [
4.813,
3.072,
2.296
],
"RecordedAtUtc": "2026-03-09T23:32:50.469885Z"
}max-usage-envelope
{
"ActiveTenants": 300,
"SelfHostedMultiplier": 4,
"MinExpectedConcurrentChecks": 100,
"MaxExpectedConcurrentChecks": 300000,
"FastestExpectedIntervalSeconds": 15,
"SlowestExpectedIntervalSeconds": 30,
"MinExpectedChecksPerSecond": "3.333",
"MinExpectedChecksPerDay": "288000.000",
"MinExpectedChecksPerWeek": "2016000.000",
"MaxExpectedChecksPerSecond": "20000.000",
"MaxExpectedChecksPerDay": "1728000000.000",
"MaxExpectedChecksPerWeek": "12096000000.000",
"StrongestFinitePlanCode": "business",
"StrongestFinitePlanMaxMonitors": 1000,
"StrongestFinitePlanMinIntervalSeconds": 15,
"StrongestFiniteChecksPerSecondPerTenant": "66.667",
"CloudBaselineChecksPerSecond": "20000.000",
"CloudBaselineChecksPerDay": "1728000000.000",
"CloudBaselineChecksPerWeek": "12096000000.000",
"SelfHostedProjectedChecksPerSecond": "80000.000",
"SelfHostedProjectedChecksPerDay": "6912000000.000",
"SelfHostedProjectedChecksPerWeek": "48384000000.000",
"FinitePlanEnvelope": [
{
"PlanCode": "business",
"MaxMonitors": 1000,
"MinIntervalSeconds": 15,
"ChecksPerSecondPerTenant": "66.667",
"CloudChecksPerSecond": "20000.000",
"CloudChecksPerDay": "1728000000.000",
"CloudChecksPerWeek": "12096000000.000",
"SelfHostedChecksPerSecond": "80000.000",
"SelfHostedChecksPerDay": "6912000000.000",
"SelfHostedChecksPerWeek": "48384000000.000"
},
{
"PlanCode": "team",
"MaxMonitors": 250,
"MinIntervalSeconds": 30,
"ChecksPerSecondPerTenant": "8.333",
"CloudChecksPerSecond": "2500.000",
"CloudChecksPerDay": "216000000.000",
"CloudChecksPerWeek": "1512000000.000",
"SelfHostedChecksPerSecond": "10000.000",
"SelfHostedChecksPerDay": "864000000.000",
"SelfHostedChecksPerWeek": "6048000000.000"
},
{
"PlanCode": "solo",
"MaxMonitors": 100,
"MinIntervalSeconds": 60,
"ChecksPerSecondPerTenant": "1.667",
"CloudChecksPerSecond": "500.000",
"CloudChecksPerDay": "43200000.000",
"CloudChecksPerWeek": "302400000.000",
"SelfHostedChecksPerSecond": "2000.000",
"SelfHostedChecksPerDay": "172800000.000",
"SelfHostedChecksPerWeek": "1209600000.000"
},
{
"PlanCode": "free",
"MaxMonitors": 50,
"MinIntervalSeconds": 120,
"ChecksPerSecondPerTenant": "0.417",
"CloudChecksPerSecond": "125.000",
"CloudChecksPerDay": "10800000.000",
"CloudChecksPerWeek": "75600000.000",
"SelfHostedChecksPerSecond": "500.000",
"SelfHostedChecksPerDay": "43200000.000",
"SelfHostedChecksPerWeek": "302400000.000"
}
],
"RecordedAtUtc": "2026-03-09T23:32:36.7969255Z"
}