{"name":"Cloud World Model Accuracy Benchmark","description":"Cloud World Model accuracy benchmark scores, campaign scopes, frozen engine versions, measurement dates, scoring methodology, and cwm-bench source references.","creator":{"@type":"Organization","@id":"https://www.cloudworldmodel.ai/#organization","name":"Cloud World Model","url":"https://www.cloudworldmodel.ai"},"url":"https://www.cloudworldmodel.ai/accuracy","jsonUrl":"https://www.cloudworldmodel.ai/accuracy.json","dateModified":"2026-11-11","engineVersion":"1.2.76","source":"https://github.com/canvascloudai/cwm-bench","scoring":{"errorRateScoring":{"tolerancePercentagePoints":0.01,"outsideTolerance":"max_inverse_tolerance_or_relative_accuracy"},"canonicalLeanOverallScore":93.8,"canonicalLeanScoringBasis":{"type":"measured_cwm_bench","costReference":"public_price_list"},"canonicalLeanReferenceProvenance":{"source":"owned_campaign","campaignId":"473f1339-f712-4096-96d6-3d4fc07cb427","campaignGeneratedAt":"2026-09-06T03:44:51.130Z","sourceRevision":"7416cb63ace3a7ab2e3486bb6f132a2dcb574c34","supplementalSourceRevision":"1fdfe97572ef91a270c9d39142fb28f7a10061eb","measurementSha":"7416cb63ace3a7ab2e3486bb6f132a2dcb574c34","resultsRevision":"117051b05b253b16b4b819d326bf1b15181af124","primaryRegion":"us-east-2","holdoutRegions":["us-west-2"],"fitScenarios":["idle","normal","peak"],"holdoutScenarios":["burst","later-day","second-region"],"repositoryUrl":"https://github.com/canvascloudai/cwm-bench","resultsUrl":"https://github.com/canvascloudai/cwm-bench/blob/117051b05b253b16b4b819d326bf1b15181af124/holdout/exports/473f1339.summary.md","runCodeUrl":"https://github.com/canvascloudai/cwm-bench/tree/7416cb63ace3a7ab2e3486bb6f132a2dcb574c34","supplementalRunCodeUrl":"https://github.com/canvascloudai/cwm-bench/tree/1fdfe97572ef91a270c9d39142fb28f7a10061eb","pricingUrl":"https://aws.amazon.com/ec2/pricing/on-demand/","predictionBasis":"SimulationEngine.simulateStep, seed 20240601; modeled in-VPC internal-LB P50/P95, mean of two compute-resource CPU predictions, whole-run goodput and fixed-fee price-list hourly cost (excluding modeled egress and LB variable processing).","cpuBasis":"Scored: mean application-server CPU across two app hosts; database CPU is measured separately. (2 × app + DB) / 3 is a derived diagnostic, not app CPU.","latencyBasis":"Scored: k6 in-VPC through an internal load balancer, excluding internet travel; engine latency is a modeled in-VPC LB path, not an end-to-end internet target.","costBasis":"Fixed-fee public price-list comparison, not a measured bill: scored prediction uses itemized resource costs before display rounding and excludes modeled egress and load-balancer variable data processing/ALB LCUs. The load-balancer base hourly fee remains included; other resource charges are not removed.","holdoutPolicy":"Fit idle, normal, and peak; hold out burst, later-day, and second-region runs.","note":"Canonical AWS scores owned app CPU, internal-LB in-VPC latency, goodput and error; $0.4545/hour is from the us-east-2 public price list, not a measured bill. Later-day and second-region have no separate engine prediction and do not enter the score."},"canonicalLeanScenarios":[{"scenario":{"id":"idle","label":"Idle","description":"Minimal background traffic — keep-alive checks, health probes, and occasional real requests.","trafficRps":10,"referenceMetrics":{"latencyP50Ms":2.62841,"latencyP95Ms":4.5295678,"cpuPct":0.4785260097091355,"throughputRps":8.874155959479843,"errorRatePct":0,"costPerHour":0.4545},"ownedMeasurement":{"source":"cwm-bench","campaignId":"473f1339-f712-4096-96d6-3d4fc07cb427","campaignGeneratedAt":"2026-09-06T03:44:51.130Z","runId":"473f1339-f712-4096-96d6-3d4fc07cb427:idle","sourceRevision":"7416cb63ace3a7ab2e3486bb6f132a2dcb574c34","measurementSha":"7416cb63ace3a7ab2e3486bb6f132a2dcb574c34","capturedAt":"2026-09-05T16:32:30.342Z","region":"us-east-2","split":"fit","targetRps":10,"metrics":{"goodputRps":8.874155959479843,"p50Ms":2.62841,"p95Ms":4.5295678,"p99Ms":7.23815152,"errorRatePct":0,"appCpuPct":0.4785260097091355,"dbCpuPct":3.337818548370722,"dbConnectionsMax":2},"blendedCpuPct":1.431623522596331,"notes":"Fit rung; goodput is the campaign's five-minute warmup plus fifteen-minute steady-window average."},"sources":[{"metric":"latencyP50Ms / latencyP95Ms","citation":"Owned k6 median/P95, measured in-VPC through an internal load balancer (internet travel excluded), idle fit run.","url":"https://github.com/canvascloudai/cwm-bench/blob/117051b05b253b16b4b819d326bf1b15181af124/holdout/exports/473f1339.summary.md"},{"metric":"cpuPct","citation":"Owned app-server CPU mean of two application hosts; database CPU measured separately, idle fit run.","url":"https://github.com/canvascloudai/cwm-bench/blob/117051b05b253b16b4b819d326bf1b15181af124/holdout/exports/473f1339.summary.md"},{"metric":"throughputRps / errorRatePct","citation":"Owned cwm-bench Admin Benchmarks campaign 473f1339: the canonical CRUD fit rung measured 8.874155959479843 RPS goodput and 0% errors at a 10 RPS target.","url":"https://github.com/canvascloudai/cwm-bench/blob/117051b05b253b16b4b819d326bf1b15181af124/holdout/exports/473f1339.summary.md"},{"metric":"costPerHour","citation":"AWS us-east-2 public price list, not a measured bill: ALB $0.0225/hr + 2× m5.large $0.096/hr + db.r5.large MySQL $0.24/hr = $0.4545/hr.","url":"https://aws.amazon.com/ec2/pricing/on-demand/"}]},"simulated":{"latencyP50Ms":2.5,"latencyP95Ms":4.4,"cpuPct":0.391,"throughputRps":8.873999999999999,"errorRatePct":0,"costPerHour":0.45449999999999996},"simulatedP99Ms":7.024620516766292,"metrics":[{"metric":"latencyP50Ms","label":"P50 in-VPC LB latency","unit":"ms","reference":2.63,"simulated":2.5,"deltaPct":-4.9,"accuracyPct":95.1},{"metric":"latencyP95Ms","label":"P95 in-VPC LB latency","unit":"ms","reference":4.53,"simulated":4.4,"deltaPct":-2.9,"accuracyPct":97.1},{"metric":"cpuPct","label":"App-server CPU","unit":"%","reference":0.48,"simulated":0.39,"deltaPct":-18.3,"accuracyPct":81.7},{"metric":"throughputRps","label":"Throughput","unit":"req/s","reference":8.87,"simulated":8.87,"deltaPct":0,"accuracyPct":100},{"metric":"errorRatePct","label":"Error Rate","unit":"%","reference":0,"simulated":0,"deltaPct":0,"accuracyPct":100},{"metric":"costPerHour","label":"Cost / Hour","unit":"$/hr","reference":0.4545,"simulated":0.4545,"deltaPct":0,"accuracyPct":100}],"compositeScore":94.6,"comparisonStatus":"complete"},{"scenario":{"id":"normal","label":"Normal","description":"Typical business-hours traffic with steady sustained load.","trafficRps":100,"referenceMetrics":{"latencyP50Ms":2.1881595,"latencyP95Ms":4.0283214,"cpuPct":1.7935897435897439,"throughputRps":87.62477401512365,"errorRatePct":0,"costPerHour":0.4545},"ownedMeasurement":{"source":"cwm-bench","campaignId":"473f1339-f712-4096-96d6-3d4fc07cb427","campaignGeneratedAt":"2026-09-06T03:44:51.130Z","runId":"473f1339-f712-4096-96d6-3d4fc07cb427:normal","sourceRevision":"7416cb63ace3a7ab2e3486bb6f132a2dcb574c34","measurementSha":"7416cb63ace3a7ab2e3486bb6f132a2dcb574c34","capturedAt":"2026-09-05T16:53:31Z","region":"us-east-2","split":"fit","targetRps":100,"metrics":{"goodputRps":87.62477401512365,"p50Ms":2.1881595,"p95Ms":4.0283214,"p99Ms":6.78735754,"errorRatePct":0,"appCpuPct":1.7935897435897439,"dbCpuPct":4.433338557944269,"dbConnectionsMax":8},"blendedCpuPct":2.673506015041252,"notes":"Fit rung; goodput is the campaign's five-minute warmup plus fifteen-minute steady-window average."},"sources":[{"metric":"latencyP50Ms / latencyP95Ms","citation":"Owned k6 median/P95, measured in-VPC through an internal load balancer (internet travel excluded), normal fit run.","url":"https://github.com/canvascloudai/cwm-bench/blob/117051b05b253b16b4b819d326bf1b15181af124/holdout/exports/473f1339.summary.md"},{"metric":"cpuPct","citation":"Owned app-server CPU mean of two application hosts; database CPU measured separately, normal fit run.","url":"https://github.com/canvascloudai/cwm-bench/blob/117051b05b253b16b4b819d326bf1b15181af124/holdout/exports/473f1339.summary.md"},{"metric":"throughputRps / errorRatePct","citation":"Owned cwm-bench Admin Benchmarks campaign 473f1339: the canonical CRUD fit rung measured 87.62477401512365 RPS goodput and 0% errors at a 100 RPS target.","url":"https://github.com/canvascloudai/cwm-bench/blob/117051b05b253b16b4b819d326bf1b15181af124/holdout/exports/473f1339.summary.md"},{"metric":"costPerHour","citation":"AWS us-east-2 public price list, not a measured bill: ALB $0.0225/hr + 2× m5.large $0.096/hr + db.r5.large MySQL $0.24/hr = $0.4545/hr.","url":"https://aws.amazon.com/ec2/pricing/on-demand/"}]},"simulated":{"latencyP50Ms":2.4,"latencyP95Ms":4.25,"cpuPct":1.901,"throughputRps":87.624,"errorRatePct":0,"costPerHour":0.4544999999999999},"simulatedP99Ms":7.048933018961295,"metrics":[{"metric":"latencyP50Ms","label":"P50 in-VPC LB latency","unit":"ms","reference":2.19,"simulated":2.4,"deltaPct":9.7,"accuracyPct":90.3},{"metric":"latencyP95Ms","label":"P95 in-VPC LB latency","unit":"ms","reference":4.03,"simulated":4.25,"deltaPct":5.5,"accuracyPct":94.5},{"metric":"cpuPct","label":"App-server CPU","unit":"%","reference":1.79,"simulated":1.9,"deltaPct":6,"accuracyPct":94},{"metric":"throughputRps","label":"Throughput","unit":"req/s","reference":87.62,"simulated":87.62,"deltaPct":0,"accuracyPct":100},{"metric":"errorRatePct","label":"Error Rate","unit":"%","reference":0,"simulated":0,"deltaPct":0,"accuracyPct":100},{"metric":"costPerHour","label":"Cost / Hour","unit":"$/hr","reference":0.4545,"simulated":0.4545,"deltaPct":0,"accuracyPct":100}],"compositeScore":95.5,"comparisonStatus":"complete"},{"scenario":{"id":"peak","label":"Peak","description":"Peak business load — product launch, end-of-day batch, or marketing campaign spike.","trafficRps":500,"referenceMetrics":{"latencyP50Ms":1.99503,"latencyP95Ms":3.75995775,"cpuPct":8.63,"throughputRps":437.62281962650434,"errorRatePct":0,"costPerHour":0.4545},"ownedMeasurement":{"source":"cwm-bench","campaignId":"473f1339-f712-4096-96d6-3d4fc07cb427","campaignGeneratedAt":"2026-09-06T03:44:51.130Z","runId":"473f1339-f712-4096-96d6-3d4fc07cb427:peak","sourceRevision":"7416cb63ace3a7ab2e3486bb6f132a2dcb574c34","measurementSha":"7416cb63ace3a7ab2e3486bb6f132a2dcb574c34","capturedAt":"2026-09-05T17:14:51Z","region":"us-east-2","split":"fit","targetRps":500,"metrics":{"goodputRps":437.62281962650434,"p50Ms":1.99503,"p95Ms":3.75995775,"p99Ms":7.20503306,"errorRatePct":0,"appCpuPct":8.63,"dbCpuPct":8.22,"dbConnectionsMax":43},"blendedCpuPct":8.493333333333334,"notes":"Fit rung; app CPU and DB CPU were reported as approximate campaign values."},"sources":[{"metric":"latencyP50Ms / latencyP95Ms","citation":"Owned k6 median/P95, measured in-VPC through an internal load balancer (internet travel excluded), peak fit run.","url":"https://github.com/canvascloudai/cwm-bench/blob/117051b05b253b16b4b819d326bf1b15181af124/holdout/exports/473f1339.summary.md"},{"metric":"cpuPct","citation":"Owned approximate app-server CPU; database CPU measured separately, peak fit run.","url":"https://github.com/canvascloudai/cwm-bench/blob/117051b05b253b16b4b819d326bf1b15181af124/holdout/exports/473f1339.summary.md"},{"metric":"throughputRps / errorRatePct","citation":"Owned cwm-bench Admin Benchmarks campaign 473f1339: the canonical CRUD fit rung measured 437.62281962650434 RPS goodput and 0% errors at a 500 RPS target.","url":"https://github.com/canvascloudai/cwm-bench/blob/117051b05b253b16b4b819d326bf1b15181af124/holdout/exports/473f1339.summary.md"},{"metric":"costPerHour","citation":"AWS us-east-2 public price list, not a measured bill: ALB $0.0225/hr + 2× m5.large $0.096/hr + db.r5.large MySQL $0.24/hr = $0.4545/hr.","url":"https://aws.amazon.com/ec2/pricing/on-demand/"}]},"simulated":{"latencyP50Ms":2,"latencyP95Ms":3.75,"cpuPct":8.61,"throughputRps":437.622913,"errorRatePct":0,"costPerHour":0.4544999999999999},"simulatedP99Ms":7.156988584272416,"metrics":[{"metric":"latencyP50Ms","label":"P50 in-VPC LB latency","unit":"ms","reference":2,"simulated":2,"deltaPct":0.2,"accuracyPct":99.8},{"metric":"latencyP95Ms","label":"P95 in-VPC LB latency","unit":"ms","reference":3.76,"simulated":3.75,"deltaPct":-0.3,"accuracyPct":99.7},{"metric":"cpuPct","label":"App-server CPU","unit":"%","reference":8.63,"simulated":8.61,"deltaPct":-0.2,"accuracyPct":99.8},{"metric":"throughputRps","label":"Throughput","unit":"req/s","reference":437.62,"simulated":437.62,"deltaPct":0,"accuracyPct":100},{"metric":"errorRatePct","label":"Error Rate","unit":"%","reference":0,"simulated":0,"deltaPct":0,"accuracyPct":100},{"metric":"costPerHour","label":"Cost / Hour","unit":"$/hr","reference":0.4545,"simulated":0.4545,"deltaPct":0,"accuracyPct":100}],"compositeScore":99.8,"comparisonStatus":"complete"},{"scenario":{"id":"burst","label":"Burst","description":"Traffic burst exceeding normal peak — flash sale, viral event, or coordinated load test.","trafficRps":1000,"referenceMetrics":{"latencyP50Ms":1.9312195,"latencyP95Ms":3.71435975,"cpuPct":13.36,"throughputRps":875.1162528375527,"errorRatePct":0.00009522521716110772,"costPerHour":0.4545},"ownedMeasurement":{"source":"cwm-bench","campaignId":"473f1339-f712-4096-96d6-3d4fc07cb427","campaignGeneratedAt":"2026-09-06T03:44:51.130Z","runId":"473f1339-f712-4096-96d6-3d4fc07cb427:burst","sourceRevision":"7416cb63ace3a7ab2e3486bb6f132a2dcb574c34","measurementSha":"7416cb63ace3a7ab2e3486bb6f132a2dcb574c34","capturedAt":"2026-09-05T17:35:55Z","region":"us-east-2","split":"holdout","targetRps":1000,"metrics":{"goodputRps":875.1162528375527,"p50Ms":1.9312195,"p95Ms":3.71435975,"p99Ms":7.01741024,"errorRatePct":0.00009522521716110772,"appCpuPct":13.36,"dbCpuPct":10.91,"dbConnectionsMax":55},"blendedCpuPct":12.543333333333331,"notes":"Holdout rung; one failure in 1,050,142 requests (raw fraction 9.522521716110773e-07, or 0.0000952%). Goodput includes a five-minute ramp and fifteen-minute steady window; no dropped requests are implied."},"sources":[{"metric":"latencyP50Ms / latencyP95Ms","citation":"Owned k6 median/P95, measured in-VPC through an internal load balancer (internet travel excluded), burst holdout run.","url":"https://github.com/canvascloudai/cwm-bench/blob/117051b05b253b16b4b819d326bf1b15181af124/holdout/exports/473f1339.summary.md"},{"metric":"cpuPct","citation":"Owned approximate app-server CPU; database CPU measured separately, burst holdout run.","url":"https://github.com/canvascloudai/cwm-bench/blob/117051b05b253b16b4b819d326bf1b15181af124/holdout/exports/473f1339.summary.md"},{"metric":"throughputRps / errorRatePct","citation":"Owned Burst holdout measured 875.1162528375527 RPS whole-run goodput and one failure in 1,050,142 (0.0000952%). Five-minute ramp plus fifteen-minute steady window; no dropped requests implied.","url":"https://github.com/canvascloudai/cwm-bench/blob/117051b05b253b16b4b819d326bf1b15181af124/holdout/exports/473f1339.summary.md"},{"metric":"costPerHour","citation":"AWS us-east-2 public price list, not a measured bill: ALB $0.0225/hr + 2× m5.large $0.096/hr + db.r5.large MySQL $0.24/hr = $0.4545/hr.","url":"https://aws.amazon.com/ec2/pricing/on-demand/"}]},"simulated":{"latencyP50Ms":1.45,"latencyP95Ms":3.1,"cpuPct":16.997,"throughputRps":875.12,"errorRatePct":0,"costPerHour":0.45449999999999996},"simulatedP99Ms":7.292058040911319,"metrics":[{"metric":"latencyP50Ms","label":"P50 in-VPC LB latency","unit":"ms","reference":1.93,"simulated":1.45,"deltaPct":-24.9,"accuracyPct":75.1},{"metric":"latencyP95Ms","label":"P95 in-VPC LB latency","unit":"ms","reference":3.71,"simulated":3.1,"deltaPct":-16.5,"accuracyPct":83.5},{"metric":"cpuPct","label":"App-server CPU","unit":"%","reference":13.36,"simulated":17,"deltaPct":27.2,"accuracyPct":72.8},{"metric":"throughputRps","label":"Throughput","unit":"req/s","reference":875.12,"simulated":875.12,"deltaPct":0,"accuracyPct":100},{"metric":"errorRatePct","label":"Error Rate","unit":"%","reference":0.000095225,"simulated":0,"deltaPct":-100,"accuracyPct":100},{"metric":"costPerHour","label":"Cost / Hour","unit":"$/hr","reference":0.4545,"simulated":0.4545,"deltaPct":0,"accuracyPct":100}],"compositeScore":85.5,"comparisonStatus":"complete"}],"canonicalProviderBases":{"aws":{"type":"measured_cwm_bench","costReference":"public_price_list","label":"Measured (CWM Bench)"},"gcp":{"type":"documentation_reference","costReference":"documentation_reference","label":"Documentation reference (not yet measured)"},"azure":{"type":"documentation_reference","costReference":"documentation_reference","label":"Documentation reference (not yet measured)"},"oci":{"type":"documentation_reference","costReference":"documentation_reference","label":"Documentation reference (not yet measured)"},"digitalocean":{"type":"documentation_reference","costReference":"documentation_reference","label":"Documentation reference (not yet measured)"}}},"campaigns":{"mysqlPreparedStatementTrajectoryV3":{"campaignId":"typical-dbsat-pscap24630-2x-v1","modelVersion":"mysql-prepared-statement-trajectory-v3","sourceRevision":"f0e1c2286f6f9b57f34021a5185f70da49e28d6c","measurementDates":["2026-10-27","2026-10-28"],"scope":"2× m5.large + db.r5.large, MySQL 8.0, max_connections 500, max_prepared_stmt_count 24,630; measured rung starts and offered-workload schedule.","claim":"Given the measured rung start times, CWM predicted the load step in which MySQL first hits its prepared-statement limit, for this topology, cap and schedule. The cap and schedule were chosen using earlier development runs.","resultsUrl":"https://github.com/canvascloudai/cwm-bench/blob/f0e1c2286f6f9b57f34021a5185f70da49e28d6c/typical/dbsat-pscap24630-2x-v1/RESULTS.md","preregistrationUrl":"https://github.com/canvascloudai/cwm-bench/blob/f0e1c2286f6f9b57f34021a5185f70da49e28d6c/typical/dbsat-pscap24630-2x-v1/PREREGISTRATION.md","evidenceUrl":"https://github.com/canvascloudai/cwm-bench/tree/f0e1c2286f6f9b57f34021a5185f70da49e28d6c/typical/dbsat-pscap24630-2x-v1/results/holdout-20261027","repetitions":[{"rep":"r1","predictedRungRps":700,"observedFirst1461RungRps":700,"forecastSecondsIntoRung":720,"firstObserved1461SecondsIntoRung":337,"onsetMinusForecastSeconds":-383},{"rep":"r2","predictedRungRps":700,"observedFirst1461RungRps":700,"forecastSecondsIntoRung":645,"firstObserved1461SecondsIntoRung":359,"onsetMinusForecastSeconds":-286}],"limitations":"The cap was hit 383 s and 286 s earlier than forecast: the forecast was late in both repetitions. First-observed poll times are not exact continuous onset times. Timing error was descriptive, not part of the rung verdict. Conditional rung validation only, not validation of exact timing, other topologies, CPU, latency or HTTP errors. Other configurations remain extrapolations. This evidence does not contribute to any aggregate accuracy score.","usageUrl":"/docs/mysql-prepared-statement-forecast.md"},"canonicalLean":{"campaignId":"473f1339-f712-4096-96d6-3d4fc07cb427","measurementDate":"2026-09-06T03:44:51.130Z","sourceRevision":"7416cb63ace3a7ab2e3486bb6f132a2dcb574c34","resultsUrl":"https://github.com/canvascloudai/cwm-bench/blob/117051b05b253b16b4b819d326bf1b15181af124/holdout/exports/473f1339.summary.md","repositoryUrl":"https://github.com/canvascloudai/cwm-bench","runCodeUrl":"https://github.com/canvascloudai/cwm-bench/tree/7416cb63ace3a7ab2e3486bb6f132a2dcb574c34","engineVersion":"1.2.76","scope":{"name":"AWS 3-Tier Web App","description":"Standard three-tier web application: Application Load Balancer distributing traffic across two m5.large EC2 instances backed by a single db.r5.large RDS MySQL instance (Single-AZ, us-east-2).","provider":"aws","region":"us-east-2","referenceRegion":"us-east-2","components":[{"role":"Load Balancer","resourceType":"network","instanceType":"ALB (Application Load Balancer)","count":1,"officialHourlyCost":0.0225,"notes":"AWS Application Load Balancer base charge (us-east-2). Scales automatically."},{"role":"Web / App Server","resourceType":"compute","instanceType":"m5.large","count":2,"officialHourlyCost":0.096,"notes":"General-purpose instance: 2 vCPU, 8 GiB RAM, up to 10 Gbps network bandwidth."},{"role":"Database","resourceType":"database","instanceType":"db.r5.large (MySQL 8.0, Single-AZ)","count":1,"officialHourlyCost":0.24,"notes":"Memory-optimized: 2 vCPU, 16 GiB RAM. gp2 storage with 3,000 IOPS burst. ~500 max connections."}]},"score":93.8},"typicalV1":{"campaignId":"typical-v1-20260927c","measurementDate":"2026-09-27","predictionsFrozenOnEngine":"1.2.5","predictionsReproducedOnEngine":"1.2.13","scope":{"campaigns":1,"days":1,"region":"us-east-2","attemptsPerRung":1,"engineVersion":"1.2.3","seed":20240601,"predictionStep":6,"predictionsFrozenBeforeMeasurement":true,"measuredCost":false,"topology":"Internal ALB + 2 × m5.large + db.r5.large MySQL 8.0; two workers and pool 250 per server"},"reportUrl":"https://github.com/canvascloudai/cwm-bench/blob/main/typical/REPORT.md","summaryUrl":"https://github.com/canvascloudai/cwm-bench/blob/main/holdout/exports/typical-v1-20260927c.summary.md","scoresUrl":"https://github.com/canvascloudai/cwm-bench/blob/main/typical/scores.json","predictionsUrl":"https://github.com/canvascloudai/cwm-bench/blob/main/typical/PREDICTIONS.md","beforeFit":{"targetRps":300,"split":"holdout","score":24.3,"scoreWithoutCost":27,"metrics":[{"metric":"p50","label":"P50","unit":"ms","predicted":17,"measured":4.621661,"measuredSource":"owned_measurement","score":0,"weight":0.2,"weightedContribution":0},{"metric":"p95","label":"P95","unit":"ms","predicted":42,"measured":26.9176766,"measuredSource":"owned_measurement","score":44,"weight":0.25,"weightedContribution":11},{"metric":"appCpu","label":"App CPU","unit":"%","predicted":59.553,"measured":15.303205128205128,"measuredSource":"owned_measurement","score":0,"weight":0.2,"weightedContribution":0},{"metric":"throughput","label":"Throughput","unit":"RPS","predicted":299,"measured":262.60872607647724,"measuredSource":"owned_measurement","score":86.1,"weight":0.15,"weightedContribution":12.9},{"metric":"errorRate","label":"Error rate","unit":"%","predicted":0.284,"measured":0.0044423431456231816,"measuredSource":"owned_measurement","score":3.6,"weight":0.1,"weightedContribution":0.4},{"metric":"cost","label":"Cost","unit":"USD/hour","predicted":0.94,"measured":0.4545,"measuredSource":"list_price_reference","score":0,"weight":0.1,"weightedContribution":0}],"unweightedP99":{"predicted":236,"measured":97.54310300000054,"unit":"ms","score":0},"throughputBasis":"Measured throughput is whole-run goodput averaged over the 5-minute warmup plus 15-minute steady window; engine throughput is a post-step point rate, not a duration-weighted average.","costBasis":"Cost was not measured. Score uses the list-price reference of USD 0.4545 per hour (ALB 0.0225 + 2 × 0.096 + 0.24, no egress, no generator). Reference, not a bill.","steadyWindowAlternative":{"scored":false,"throughputRps":300,"throughputScore":99.7,"totalScore":26.3}},"afterFit":{"targetRps":300,"split":"independent holdout","engineVersion":"1.2.20","calibrationId":"aws-crud/typical:typical-v1-20260927c:6aa574d7ff9d3080b88b221bcd59f7d218ae37f0","fitRungs":["typical-fit-20","typical-fit-100","typical-fit-200"],"score":84.2,"scoreWithoutCost":82.4,"metrics":[{"metric":"p50","label":"P50","unit":"ms","predicted":4.165220860655737,"measured":4.621661,"measuredSource":"owned_measurement","score":90.1,"weight":0.2,"weightedContribution":18},{"metric":"p95","label":"P95","unit":"ms","predicted":12.78397378032783,"measured":26.9176766,"measuredSource":"owned_measurement","score":47.5,"weight":0.25,"weightedContribution":11.9},{"metric":"appCpu","label":"App CPU","unit":"%","predicted":14.74,"measured":15.303205128205128,"measuredSource":"owned_measurement","score":96.3,"weight":0.2,"weightedContribution":19.3},{"metric":"throughput","label":"Throughput","unit":"RPS","predicted":262.604014,"measured":262.60872607647724,"measuredSource":"owned_measurement","score":100,"weight":0.15,"weightedContribution":15},{"metric":"errorRate","label":"Error rate","unit":"%","predicted":0,"measured":0.0044423431456231816,"measuredSource":"owned_measurement","score":100,"weight":0.1,"weightedContribution":10},{"metric":"cost","label":"Cost","unit":"USD/hour","predicted":0.454581,"measured":0.4545,"measuredSource":"list_price_reference","score":100,"weight":0.1,"weightedContribution":10}],"throughputBasis":"Measured throughput is whole-run goodput averaged over the 5-minute warmup plus 15-minute steady window; engine throughput is a post-step point rate, not a duration-weighted average. The post-fit model was fit to whole-run goodput at 20, 100, and 200 RPS.","costBasis":"Cost was not measured. USD 0.4545/hour is a fixed-fee public list-price reference, not a bill. The frozen post-fit scored prediction predates modeled LB variable processing. It excludes modeled egress; the historical before-fit prediction included it, so their cost scores are not on identical bases."},"dayAndRegionHoldouts":{"campaignId":"typical-holdouts-v1-20261006","reportUrl":"https://github.com/canvascloudai/cwm-bench/blob/main/typical/REPORT.md#day-and-region-holdouts-typical-holdouts-v1-20261006","comparisonBasis":"page cost basis","original":{"campaignId":"typical-v1-20260927c","targetRps":300,"score":84.2,"scoreWithoutCost":82.4,"p95Ms":26.9},"observed":[{"id":"later-day","label":"Later-day","region":"us-east-2","targetRps":300,"score":76,"scoreWithoutCost":73.3,"p95Ms":51.1},{"id":"second-region","label":"Second region","region":"us-west-2","targetRps":300,"score":81.1,"scoreWithoutCost":79,"p95Ms":32.9}],"repeatability100Rps":{"targetRps":100,"scores":[95.6,91.4]}},"extrapolatedDiagnostic":{"targetRps":500,"label":"diagnostic extrapolation beyond the calibrated and validated range (fit at 20, 100, and 200 RPS; validated at 300 RPS), compared with a measured 500 RPS diagnostic run; excluded from headline accuracy and not a validated saturation finding","engineVersion":"1.2.20","calibrationId":"aws-crud/typical:typical-v1-20260927c:6aa574d7ff9d3080b88b221bcd59f7d218ae37f0","fitRungs":["typical-fit-20","typical-fit-100","typical-fit-200"],"score":64.1,"scoreWithoutCost":60.1,"metrics":[{"metric":"p50","label":"P50","unit":"ms","predicted":4.165220860655737,"measured":5.187581,"measuredSource":"owned_measurement","score":80.3,"weight":0.2,"weightedContribution":16.1},{"metric":"p95","label":"P95","unit":"ms","predicted":16.333451173770428,"measured":88.189861,"measuredSource":"owned_measurement","score":18.5,"weight":0.25,"weightedContribution":4.6},{"metric":"appCpu","label":"App CPU","unit":"%","predicted":24.264,"measured":26.34166695034375,"measuredSource":"owned_measurement","score":92.1,"weight":0.2,"weightedContribution":18.4},{"metric":"throughput","label":"Throughput","unit":"RPS","predicted":437.588385,"measured":437.5769855710571,"measuredSource":"owned_measurement","score":100,"weight":0.15,"weightedContribution":15},{"metric":"errorRate","label":"Error rate","unit":"%","predicted":12.482,"measured":0.010282815511055931,"measuredSource":"owned_measurement","score":0.1,"weight":0.1,"weightedContribution":0},{"metric":"cost","label":"Cost","unit":"USD/hour","predicted":0.4545,"measured":0.4545,"measuredSource":"list_price_reference","score":100,"weight":0.1,"weightedContribution":10}],"throughputBasis":"Measured throughput is whole-run goodput averaged over the 5-minute warmup plus 15-minute steady window; engine throughput is a post-step point rate, not a duration-weighted average. The post-fit model was fit to whole-run goodput at 20, 100, and 200 RPS.","costBasis":"USD 0.4545/hour is a fixed-fee public list-price reference, not a measured bill. Fixed-fee public price-list comparison, not a measured bill: scored prediction uses itemized resource costs before display rounding and excludes modeled egress and load-balancer variable data processing/ALB LCUs. The load-balancer base hourly fee remains included; other resource charges are not removed."}},"typicalScaleV1":{"campaignId":"typical-scale-v1","measurementDate":"2026-10-06","sourceUrl":"https://github.com/canvascloudai/cwm-bench/pull/18","reportUrl":"https://github.com/canvascloudai/cwm-bench/blob/3b6708cee5b00f61254b0027c215ba0e1f389170/typical/REPORT.md#app-server-count-change-typical-scale-v1-20261006","predictionsFrozenOnEngine":"1.2.14","scope":"us-east-2, m5.large app servers (1–3), 100–300 total RPS, internal load balancer, and db.r5.large MySQL","results":[{"id":"1x-100","servers":1,"rps":100,"measuredP95":"9.1","predictedP95":"11.0","score":"89.1","low":"81.5","high":"93.6"},{"id":"1x-200","servers":1,"rps":200,"measuredP95":"37.5","predictedP95":"14.6","score":"82.1","low":"79.8","high":"83.8"},{"id":"1x-300","servers":1,"rps":300,"measuredP95":"95.0","predictedP95":"18.1","score":"74.5","low":"74.0","high":"75.0"},{"id":"2x-100","servers":2,"rps":100,"measuredP95":"8.4","predictedP95":"9.2","score":"95.7","low":"94.4","high":"97.1"},{"id":"2x-200","servers":2,"rps":200,"measuredP95":"15.1","predictedP95":"11.0","score":"89.8","low":"81.1","high":"95.8"},{"id":"2x-300","servers":2,"rps":300,"measuredP95":"41.3","predictedP95":"12.8","score":"77.7","low":"77.5","high":"80.3"},{"id":"3x-100","servers":3,"rps":100,"measuredP95":"8.7","predictedP95":"8.6","score":"98.4","low":"96.6","high":"98.4"},{"id":"3x-200","servers":3,"rps":200,"measuredP95":"12.1","predictedP95":"9.8","score":"94.7","low":"90.4","high":"94.9"},{"id":"3x-300","servers":3,"rps":300,"measuredP95":"20.7","predictedP95":"11.0","score":"86.8","low":"84.9","high":"89.2"}]},"typicalP95V1":{"campaignId":"typical-p95-v1-20261010","measurementDates":["2026-10-10","2026-10-12"],"sourceUrl":"https://github.com/canvascloudai/cwm-bench/pull/21","reportUrl":"https://github.com/canvascloudai/cwm-bench/blob/main/typical/REPORT.md#p95-two-input-model-holdout-typical-p95-v1-20261010","preregistrationUrl":"https://github.com/canvascloudai/cwm-bench/blob/main/typical/p95-v1/PREREGISTRATION.md","predictionsFrozenOnEngine":"1.2.18","scope":"us-east-2, m5.large app servers (1–3), 100–300 total RPS, internal load balancer, and db.r5.large MySQL","medianP95Score":87.6,"results":[{"id":"1x-100","servers":1,"rps":100,"measured":"8.6","candidate":"9.6","engine":"11.0","candidateScore":"89.0","engineScore":"72.6"},{"id":"1x-200","servers":1,"rps":200,"measured":"40.3","candidate":"32.0","engine":"14.6","candidateScore":"79.3","engineScore":"36.1"},{"id":"1x-300","servers":1,"rps":300,"measured":"94.7","candidate":"106.5","engine":"18.1","candidateScore":"87.6","engineScore":"19.1"},{"id":"2x-100","servers":2,"rps":100,"measured":"9.2","candidate":"8.0","engine":"9.2","candidateScore":"87.1","engineScore":"99.5"},{"id":"2x-200","servers":2,"rps":200,"measured":"15.7","candidate":"15.0","engine":"11.0","candidateScore":"95.7","engineScore":"70.2"},{"id":"2x-300","servers":2,"rps":300,"measured":"38.2","candidate":"34.3","engine":"12.8","candidateScore":"89.7","engineScore":"33.5"},{"id":"3x-100","servers":3,"rps":100,"measured":"7.9","candidate":"8.0","engine":"8.6","candidateScore":"99.1","engineScore":"91.0"},{"id":"3x-200","servers":3,"rps":200,"measured":"9.0","candidate":"11.7","engine":"9.8","candidateScore":"71.2","engineScore":"91.4"},{"id":"3x-300","servers":3,"rps":300,"measured":"17.9","candidate":"23.5","engine":"11.0","candidateScore":"68.8","engineScore":"61.5"}]},"mysqlCpuServicePoolV4":{"campaignId":"typical-dbsat-cpu-2x-m5xl-v1","measurementDates":["2026-10-25"],"predictionsFrozenOnModel":"mysql-cpu-service-pool-v4","sourceRevision":"8ec726ec4e26fe4092f0558399138d88ee053034","sourceUrl":"https://github.com/canvascloudai/cwm-bench/tree/8ec726ec4e26fe4092f0558399138d88ee053034/typical/dbsat-cpu-2x-m5xl-v1","reportUrl":"https://github.com/canvascloudai/cwm-bench/blob/8ec726ec4e26fe4092f0558399138d88ee053034/typical/dbsat-cpu-2x-m5xl-v1/RESULTS.md","preregistrationUrl":"https://github.com/canvascloudai/cwm-bench/blob/8ec726ec4e26fe4092f0558399138d88ee053034/typical/dbsat-cpu-2x-m5xl-v1/PREREGISTRATION.md","scope":"AWS us-east-2, two fixed m5.large app servers, one db.m5.xlarge MySQL 8.0 database, typical workload, 300–1,500 RPS, 500 configured database connections","databaseSize":"db.m5.xlarge","rps":1500,"measuredRepetitions":[{"id":"rep-1","dbCpuPercent":96.2,"errorRatePct":15.4,"httpP95Ms":1017},{"id":"rep-2","dbCpuPercent":93.8,"errorRatePct":9,"httpP95Ms":934}],"frozenPrediction":{"dbCpuPercent":99.7,"errorRatePct":12.22656385145596,"httpP95Ms":1282.8409989671352,"dominantFailureClass":"queue_full","dominantFailureSharePct":96.4873931423609},"conditionalFirstBreach":{"rps":1500,"status":"conditional_stationary_breach","criterion":"error","dominantClass":"queue_full","timing":"The model predicts errors cross the threshold at 1,500 RPS once traffic holds steady for 3 minutes; this is a projection, not a measured timeline."},"historicalDbR5XlargeH0":{"retainedSeparately":true,"predictionOnly":false,"evidenceStatus":"measured","hypothesisOutcome":"H0","campaignId":"typical-dbsat-cpu-2x-r5xl-v1","databaseSize":"db.r5.xlarge","calibrationVersion":"mysql-cpu-service-pool-v3","sourceRevision":"c6bd50e4e8f6c263edc6b9bf2c5f9e421eb11112","sourceUrl":"https://github.com/canvascloudai/cwm-bench/tree/c6bd50e4e8f6c263edc6b9bf2c5f9e421eb11112/typical/dbsat-cpu-2x-r5xl-v1","resultsUrl":"https://github.com/canvascloudai/cwm-bench/blob/main/typical/dbsat-cpu-2x-r5xl-v1/RESULTS.md","lastRungRps":1500,"measuredOutcome":{"firstBreachRps":1500,"criterion":"error","dominantFailureClass":"queue_full","runs":[{"id":"run-1","verdict":"H0","dbCpuMeanPercent":93.7,"dbCpuPeakPercent":95.6,"errorRatePct":10.1,"httpP95Ms":956,"queueFullFailureSharePct":81.6},{"id":"run-2","verdict":"H0","dbCpuMeanPercent":96,"dbCpuPeakPercent":96.7,"errorRatePct":17.14,"httpP95Ms":1078,"queueFullFailureSharePct":79.9}]},"frozenPrediction":{"firstBreach":"no_breach_on_tested_ladder","dbCpuPercent":75.86904978522632,"errorRatePct":0.18095016995165655,"httpP95Ms":317.092763031571,"maxThreadsConnectedProxy":503}},"observedOutcome":{"status":"passed_db_family_transfer","firstBreachRps":1500,"criterion":"error","mechanism":"DB-limited","dominantFailureClass":"queue_full","queueFullFailureSharePctRange":[76,79]},"caveats":["v4 overpredicts errors below the breach.","At 1,300 RPS, measured headroom was lower than the frozen model predicted."]}},"providerBenchmarks":{"aws":{"architecture":{"name":"AWS 3-Tier Web App","description":"Standard three-tier web application: Application Load Balancer distributing traffic across two m5.large EC2 instances backed by a single db.r5.large RDS MySQL instance (Single-AZ, us-east-2).","provider":"aws","region":"us-east-2","referenceRegion":"us-east-2","components":[{"role":"Load Balancer","resourceType":"network","instanceType":"ALB (Application Load Balancer)","count":1,"officialHourlyCost":0.0225,"notes":"AWS Application Load Balancer base charge (us-east-2). Scales automatically."},{"role":"Web / App Server","resourceType":"compute","instanceType":"m5.large","count":2,"officialHourlyCost":0.096,"notes":"General-purpose instance: 2 vCPU, 8 GiB RAM, up to 10 Gbps network bandwidth."},{"role":"Database","resourceType":"database","instanceType":"db.r5.large (MySQL 8.0, Single-AZ)","count":1,"officialHourlyCost":0.24,"notes":"Memory-optimized: 2 vCPU, 16 GiB RAM. gp2 storage with 3,000 IOPS burst. ~500 max connections."}]},"typicalOwnedHoldout":{"appWeight":"typical","source":"owned_cwm_bench","campaignId":"typical-v1-20260927c","measurementCommit":"6aa574d7ff9d3080b88b221bcd59f7d218ae37f0","reportUrl":"https://github.com/canvascloudai/cwm-bench/blob/main/typical/REPORT.md","summaryUrl":"https://github.com/canvascloudai/cwm-bench/blob/main/holdout/exports/typical-v1-20260927c.summary.md","scoresUrl":"https://github.com/canvascloudai/cwm-bench/blob/main/typical/scores.json","predictionsUrl":"https://github.com/canvascloudai/cwm-bench/blob/main/typical/PREDICTIONS.md","scope":{"campaigns":1,"days":1,"region":"us-east-2","attemptsPerRung":1,"engineVersion":"1.2.3","seed":20240601,"predictionStep":6,"predictionsFrozenBeforeMeasurement":true,"measuredCost":false,"topology":"Internal ALB + 2 × m5.large + db.r5.large MySQL 8.0; two workers and pool 250 per server"},"beforeFit":{"targetRps":300,"split":"holdout","score":24.3,"scoreWithoutCost":27,"metrics":[{"metric":"p50","label":"P50","unit":"ms","predicted":17,"measured":4.621661,"measuredSource":"owned_measurement","score":0,"weight":0.2,"weightedContribution":0},{"metric":"p95","label":"P95","unit":"ms","predicted":42,"measured":26.9176766,"measuredSource":"owned_measurement","score":44,"weight":0.25,"weightedContribution":11},{"metric":"appCpu","label":"App CPU","unit":"%","predicted":59.553,"measured":15.303205128205128,"measuredSource":"owned_measurement","score":0,"weight":0.2,"weightedContribution":0},{"metric":"throughput","label":"Throughput","unit":"RPS","predicted":299,"measured":262.60872607647724,"measuredSource":"owned_measurement","score":86.1,"weight":0.15,"weightedContribution":12.9},{"metric":"errorRate","label":"Error rate","unit":"%","predicted":0.284,"measured":0.0044423431456231816,"measuredSource":"owned_measurement","score":3.6,"weight":0.1,"weightedContribution":0.4},{"metric":"cost","label":"Cost","unit":"USD/hour","predicted":0.94,"measured":0.4545,"measuredSource":"list_price_reference","score":0,"weight":0.1,"weightedContribution":0}],"unweightedP99":{"predicted":236,"measured":97.54310300000054,"unit":"ms","score":0},"throughputBasis":"Measured throughput is whole-run goodput averaged over the 5-minute warmup plus 15-minute steady window; engine throughput is a post-step point rate, not a duration-weighted average.","costBasis":"Cost was not measured. Score uses the list-price reference of USD 0.4545 per hour (ALB 0.0225 + 2 × 0.096 + 0.24, no egress, no generator). Reference, not a bill.","steadyWindowAlternative":{"scored":false,"throughputRps":300,"throughputScore":99.7,"totalScore":26.3}},"dayAndRegionHoldouts":{"campaignId":"typical-holdouts-v1-20261006","reportUrl":"https://github.com/canvascloudai/cwm-bench/blob/main/typical/REPORT.md#day-and-region-holdouts-typical-holdouts-v1-20261006","comparisonBasis":"page cost basis","original":{"campaignId":"typical-v1-20260927c","targetRps":300,"score":84.2,"scoreWithoutCost":82.4,"p95Ms":26.9},"observed":[{"id":"later-day","label":"Later-day","region":"us-east-2","targetRps":300,"score":76,"scoreWithoutCost":73.3,"p95Ms":51.1},{"id":"second-region","label":"Second region","region":"us-west-2","targetRps":300,"score":81.1,"scoreWithoutCost":79,"p95Ms":32.9}],"repeatability100Rps":{"targetRps":100,"scores":[95.6,91.4]}},"context":[{"rung":"fit-20","role":"fit_data_not_holdout","score":23,"scoreWithoutCost":15.3},{"rung":"fit-100","role":"fit_data_not_holdout","score":19.8,"scoreWithoutCost":14.9},{"rung":"fit-200","role":"fit_data_not_holdout","score":16.1,"scoreWithoutCost":14.8},{"rung":"saturation-500","role":"diagnostic_only","score":25.6,"scoreWithoutCost":28.5}],"extrapolatedDiagnostic":{"targetRps":500,"label":"diagnostic extrapolation beyond the calibrated and validated range (fit at 20, 100, and 200 RPS; validated at 300 RPS), compared with a measured 500 RPS diagnostic run; excluded from headline accuracy and not a validated saturation finding","engineVersion":"1.2.20","calibrationId":"aws-crud/typical:typical-v1-20260927c:6aa574d7ff9d3080b88b221bcd59f7d218ae37f0","fitRungs":["typical-fit-20","typical-fit-100","typical-fit-200"],"score":64.1,"scoreWithoutCost":60.1,"metrics":[{"metric":"p50","label":"P50","unit":"ms","predicted":4.165220860655737,"measured":5.187581,"measuredSource":"owned_measurement","score":80.3,"weight":0.2,"weightedContribution":16.1},{"metric":"p95","label":"P95","unit":"ms","predicted":16.333451173770428,"measured":88.189861,"measuredSource":"owned_measurement","score":18.5,"weight":0.25,"weightedContribution":4.6},{"metric":"appCpu","label":"App CPU","unit":"%","predicted":24.264,"measured":26.34166695034375,"measuredSource":"owned_measurement","score":92.1,"weight":0.2,"weightedContribution":18.4},{"metric":"throughput","label":"Throughput","unit":"RPS","predicted":437.588385,"measured":437.5769855710571,"measuredSource":"owned_measurement","score":100,"weight":0.15,"weightedContribution":15},{"metric":"errorRate","label":"Error rate","unit":"%","predicted":12.482,"measured":0.010282815511055931,"measuredSource":"owned_measurement","score":0.1,"weight":0.1,"weightedContribution":0},{"metric":"cost","label":"Cost","unit":"USD/hour","predicted":0.4545,"measured":0.4545,"measuredSource":"list_price_reference","score":100,"weight":0.1,"weightedContribution":10}],"throughputBasis":"Measured throughput is whole-run goodput averaged over the 5-minute warmup plus 15-minute steady window; engine throughput is a post-step point rate, not a duration-weighted average. The post-fit model was fit to whole-run goodput at 20, 100, and 200 RPS.","costBasis":"USD 0.4545/hour is a fixed-fee public list-price reference, not a measured bill. Fixed-fee public price-list comparison, not a measured bill: scored prediction uses itemized resource costs before display rounding and excludes modeled egress and load-balancer variable data processing/ALB LCUs. The load-balancer base hourly fee remains included; other resource charges are not removed."},"afterFit":{"targetRps":300,"split":"independent holdout","engineVersion":"1.2.20","calibrationId":"aws-crud/typical:typical-v1-20260927c:6aa574d7ff9d3080b88b221bcd59f7d218ae37f0","fitRungs":["typical-fit-20","typical-fit-100","typical-fit-200"],"score":84.2,"scoreWithoutCost":82.4,"metrics":[{"metric":"p50","label":"P50","unit":"ms","predicted":4.165220860655737,"measured":4.621661,"measuredSource":"owned_measurement","score":90.1,"weight":0.2,"weightedContribution":18},{"metric":"p95","label":"P95","unit":"ms","predicted":12.78397378032783,"measured":26.9176766,"measuredSource":"owned_measurement","score":47.5,"weight":0.25,"weightedContribution":11.9},{"metric":"appCpu","label":"App CPU","unit":"%","predicted":14.74,"measured":15.303205128205128,"measuredSource":"owned_measurement","score":96.3,"weight":0.2,"weightedContribution":19.3},{"metric":"throughput","label":"Throughput","unit":"RPS","predicted":262.604014,"measured":262.60872607647724,"measuredSource":"owned_measurement","score":100,"weight":0.15,"weightedContribution":15},{"metric":"errorRate","label":"Error rate","unit":"%","predicted":0,"measured":0.0044423431456231816,"measuredSource":"owned_measurement","score":100,"weight":0.1,"weightedContribution":10},{"metric":"cost","label":"Cost","unit":"USD/hour","predicted":0.454581,"measured":0.4545,"measuredSource":"list_price_reference","score":100,"weight":0.1,"weightedContribution":10}],"throughputBasis":"Measured throughput is whole-run goodput averaged over the 5-minute warmup plus 15-minute steady window; engine throughput is a post-step point rate, not a duration-weighted average. The post-fit model was fit to whole-run goodput at 20, 100, and 200 RPS.","costBasis":"Cost was not measured. USD 0.4545/hour is a fixed-fee public list-price reference, not a bill. The frozen post-fit scored prediction predates modeled LB variable processing. It excludes modeled egress; the historical before-fit prediction included it, so their cost scores are not on identical bases."}},"seed":20240601,"generatedAt":"2026-11-16T04:47:50.927Z","overallScore":93.8,"scoringBasis":{"type":"measured_cwm_bench","costReference":"public_price_list"},"errorRateScoring":{"tolerancePercentagePoints":0.01,"outsideTolerance":"max_inverse_tolerance_or_relative_accuracy"},"scoreDriverNote":"Canonical AWS compares seeded engine app-server CPU and modeled in-VPC internal-load-balancer latency with owned cwm-bench measurements. Goodput/errors are owned; billed cost uses the us-east-2 price list. CPU and latency misses are shown without engine retuning.","fidelity":{"scorecardIncluded":true,"label":"mixed","components":{"cost":"price_list","latency":"owned","cpu":"owned","throughput":"owned","errorRate":"owned"}},"referenceProvenance":{"source":"owned_campaign","campaignId":"473f1339-f712-4096-96d6-3d4fc07cb427","campaignGeneratedAt":"2026-09-06T03:44:51.130Z","sourceRevision":"7416cb63ace3a7ab2e3486bb6f132a2dcb574c34","supplementalSourceRevision":"1fdfe97572ef91a270c9d39142fb28f7a10061eb","measurementSha":"7416cb63ace3a7ab2e3486bb6f132a2dcb574c34","resultsRevision":"117051b05b253b16b4b819d326bf1b15181af124","primaryRegion":"us-east-2","holdoutRegions":["us-west-2"],"fitScenarios":["idle","normal","peak"],"holdoutScenarios":["burst","later-day","second-region"],"repositoryUrl":"https://github.com/canvascloudai/cwm-bench","resultsUrl":"https://github.com/canvascloudai/cwm-bench/blob/117051b05b253b16b4b819d326bf1b15181af124/holdout/exports/473f1339.summary.md","runCodeUrl":"https://github.com/canvascloudai/cwm-bench/tree/7416cb63ace3a7ab2e3486bb6f132a2dcb574c34","supplementalRunCodeUrl":"https://github.com/canvascloudai/cwm-bench/tree/1fdfe97572ef91a270c9d39142fb28f7a10061eb","pricingUrl":"https://aws.amazon.com/ec2/pricing/on-demand/","predictionBasis":"SimulationEngine.simulateStep, seed 20240601; modeled in-VPC internal-LB P50/P95, mean of two compute-resource CPU predictions, whole-run goodput and fixed-fee price-list hourly cost (excluding modeled egress and LB variable processing).","cpuBasis":"Scored: mean application-server CPU across two app hosts; database CPU is measured separately. (2 × app + DB) / 3 is a derived diagnostic, not app CPU.","latencyBasis":"Scored: k6 in-VPC through an internal load balancer, excluding internet travel; engine latency is a modeled in-VPC LB path, not an end-to-end internet target.","costBasis":"Fixed-fee public price-list comparison, not a measured bill: scored prediction uses itemized resource costs before display rounding and excludes modeled egress and load-balancer variable data processing/ALB LCUs. The load-balancer base hourly fee remains included; other resource charges are not removed.","holdoutPolicy":"Fit idle, normal, and peak; hold out burst, later-day, and second-region runs.","note":"Canonical AWS scores owned app CPU, internal-LB in-VPC latency, goodput and error; $0.4545/hour is from the us-east-2 public price list, not a measured bill. Later-day and second-region have no separate engine prediction and do not enter the score."},"measurementSource":"owned_campaign","enginePredictionStatus":"available","scenarios":[{"scenario":{"id":"idle","label":"Idle","description":"Minimal background traffic — keep-alive checks, health probes, and occasional real requests.","trafficRps":10,"referenceMetrics":{"latencyP50Ms":2.62841,"latencyP95Ms":4.5295678,"cpuPct":0.4785260097091355,"throughputRps":8.874155959479843,"errorRatePct":0,"costPerHour":0.4545},"ownedMeasurement":{"source":"cwm-bench","campaignId":"473f1339-f712-4096-96d6-3d4fc07cb427","campaignGeneratedAt":"2026-09-06T03:44:51.130Z","runId":"473f1339-f712-4096-96d6-3d4fc07cb427:idle","sourceRevision":"7416cb63ace3a7ab2e3486bb6f132a2dcb574c34","measurementSha":"7416cb63ace3a7ab2e3486bb6f132a2dcb574c34","capturedAt":"2026-09-05T16:32:30.342Z","region":"us-east-2","split":"fit","targetRps":10,"metrics":{"goodputRps":8.874155959479843,"p50Ms":2.62841,"p95Ms":4.5295678,"p99Ms":7.23815152,"errorRatePct":0,"appCpuPct":0.4785260097091355,"dbCpuPct":3.337818548370722,"dbConnectionsMax":2},"blendedCpuPct":1.431623522596331,"notes":"Fit rung; goodput is the campaign's five-minute warmup plus fifteen-minute steady-window average."},"sources":[{"metric":"latencyP50Ms / latencyP95Ms","citation":"Owned k6 median/P95, measured in-VPC through an internal load balancer (internet travel excluded), idle fit run.","url":"https://github.com/canvascloudai/cwm-bench/blob/117051b05b253b16b4b819d326bf1b15181af124/holdout/exports/473f1339.summary.md"},{"metric":"cpuPct","citation":"Owned app-server CPU mean of two application hosts; database CPU measured separately, idle fit run.","url":"https://github.com/canvascloudai/cwm-bench/blob/117051b05b253b16b4b819d326bf1b15181af124/holdout/exports/473f1339.summary.md"},{"metric":"throughputRps / errorRatePct","citation":"Owned cwm-bench Admin Benchmarks campaign 473f1339: the canonical CRUD fit rung measured 8.874155959479843 RPS goodput and 0% errors at a 10 RPS target.","url":"https://github.com/canvascloudai/cwm-bench/blob/117051b05b253b16b4b819d326bf1b15181af124/holdout/exports/473f1339.summary.md"},{"metric":"costPerHour","citation":"AWS us-east-2 public price list, not a measured bill: ALB $0.0225/hr + 2× m5.large $0.096/hr + db.r5.large MySQL $0.24/hr = $0.4545/hr.","url":"https://aws.amazon.com/ec2/pricing/on-demand/"}]},"simulated":{"latencyP50Ms":2.5,"latencyP95Ms":4.4,"cpuPct":0.391,"throughputRps":8.873999999999999,"errorRatePct":0,"costPerHour":0.45449999999999996},"simulatedP99Ms":7.024620516766292,"metrics":[{"metric":"latencyP50Ms","label":"P50 in-VPC LB latency","unit":"ms","reference":2.63,"simulated":2.5,"deltaPct":-4.9,"accuracyPct":95.1},{"metric":"latencyP95Ms","label":"P95 in-VPC LB latency","unit":"ms","reference":4.53,"simulated":4.4,"deltaPct":-2.9,"accuracyPct":97.1},{"metric":"cpuPct","label":"App-server CPU","unit":"%","reference":0.48,"simulated":0.39,"deltaPct":-18.3,"accuracyPct":81.7},{"metric":"throughputRps","label":"Throughput","unit":"req/s","reference":8.87,"simulated":8.87,"deltaPct":0,"accuracyPct":100},{"metric":"errorRatePct","label":"Error Rate","unit":"%","reference":0,"simulated":0,"deltaPct":0,"accuracyPct":100},{"metric":"costPerHour","label":"Cost / Hour","unit":"$/hr","reference":0.4545,"simulated":0.4545,"deltaPct":0,"accuracyPct":100}],"compositeScore":94.6,"comparisonStatus":"complete"},{"scenario":{"id":"normal","label":"Normal","description":"Typical business-hours traffic with steady sustained load.","trafficRps":100,"referenceMetrics":{"latencyP50Ms":2.1881595,"latencyP95Ms":4.0283214,"cpuPct":1.7935897435897439,"throughputRps":87.62477401512365,"errorRatePct":0,"costPerHour":0.4545},"ownedMeasurement":{"source":"cwm-bench","campaignId":"473f1339-f712-4096-96d6-3d4fc07cb427","campaignGeneratedAt":"2026-09-06T03:44:51.130Z","runId":"473f1339-f712-4096-96d6-3d4fc07cb427:normal","sourceRevision":"7416cb63ace3a7ab2e3486bb6f132a2dcb574c34","measurementSha":"7416cb63ace3a7ab2e3486bb6f132a2dcb574c34","capturedAt":"2026-09-05T16:53:31Z","region":"us-east-2","split":"fit","targetRps":100,"metrics":{"goodputRps":87.62477401512365,"p50Ms":2.1881595,"p95Ms":4.0283214,"p99Ms":6.78735754,"errorRatePct":0,"appCpuPct":1.7935897435897439,"dbCpuPct":4.433338557944269,"dbConnectionsMax":8},"blendedCpuPct":2.673506015041252,"notes":"Fit rung; goodput is the campaign's five-minute warmup plus fifteen-minute steady-window average."},"sources":[{"metric":"latencyP50Ms / latencyP95Ms","citation":"Owned k6 median/P95, measured in-VPC through an internal load balancer (internet travel excluded), normal fit run.","url":"https://github.com/canvascloudai/cwm-bench/blob/117051b05b253b16b4b819d326bf1b15181af124/holdout/exports/473f1339.summary.md"},{"metric":"cpuPct","citation":"Owned app-server CPU mean of two application hosts; database CPU measured separately, normal fit run.","url":"https://github.com/canvascloudai/cwm-bench/blob/117051b05b253b16b4b819d326bf1b15181af124/holdout/exports/473f1339.summary.md"},{"metric":"throughputRps / errorRatePct","citation":"Owned cwm-bench Admin Benchmarks campaign 473f1339: the canonical CRUD fit rung measured 87.62477401512365 RPS goodput and 0% errors at a 100 RPS target.","url":"https://github.com/canvascloudai/cwm-bench/blob/117051b05b253b16b4b819d326bf1b15181af124/holdout/exports/473f1339.summary.md"},{"metric":"costPerHour","citation":"AWS us-east-2 public price list, not a measured bill: ALB $0.0225/hr + 2× m5.large $0.096/hr + db.r5.large MySQL $0.24/hr = $0.4545/hr.","url":"https://aws.amazon.com/ec2/pricing/on-demand/"}]},"simulated":{"latencyP50Ms":2.4,"latencyP95Ms":4.25,"cpuPct":1.901,"throughputRps":87.624,"errorRatePct":0,"costPerHour":0.4544999999999999},"simulatedP99Ms":7.048933018961295,"metrics":[{"metric":"latencyP50Ms","label":"P50 in-VPC LB latency","unit":"ms","reference":2.19,"simulated":2.4,"deltaPct":9.7,"accuracyPct":90.3},{"metric":"latencyP95Ms","label":"P95 in-VPC LB latency","unit":"ms","reference":4.03,"simulated":4.25,"deltaPct":5.5,"accuracyPct":94.5},{"metric":"cpuPct","label":"App-server CPU","unit":"%","reference":1.79,"simulated":1.9,"deltaPct":6,"accuracyPct":94},{"metric":"throughputRps","label":"Throughput","unit":"req/s","reference":87.62,"simulated":87.62,"deltaPct":0,"accuracyPct":100},{"metric":"errorRatePct","label":"Error Rate","unit":"%","reference":0,"simulated":0,"deltaPct":0,"accuracyPct":100},{"metric":"costPerHour","label":"Cost / Hour","unit":"$/hr","reference":0.4545,"simulated":0.4545,"deltaPct":0,"accuracyPct":100}],"compositeScore":95.5,"comparisonStatus":"complete"},{"scenario":{"id":"peak","label":"Peak","description":"Peak business load — product launch, end-of-day batch, or marketing campaign spike.","trafficRps":500,"referenceMetrics":{"latencyP50Ms":1.99503,"latencyP95Ms":3.75995775,"cpuPct":8.63,"throughputRps":437.62281962650434,"errorRatePct":0,"costPerHour":0.4545},"ownedMeasurement":{"source":"cwm-bench","campaignId":"473f1339-f712-4096-96d6-3d4fc07cb427","campaignGeneratedAt":"2026-09-06T03:44:51.130Z","runId":"473f1339-f712-4096-96d6-3d4fc07cb427:peak","sourceRevision":"7416cb63ace3a7ab2e3486bb6f132a2dcb574c34","measurementSha":"7416cb63ace3a7ab2e3486bb6f132a2dcb574c34","capturedAt":"2026-09-05T17:14:51Z","region":"us-east-2","split":"fit","targetRps":500,"metrics":{"goodputRps":437.62281962650434,"p50Ms":1.99503,"p95Ms":3.75995775,"p99Ms":7.20503306,"errorRatePct":0,"appCpuPct":8.63,"dbCpuPct":8.22,"dbConnectionsMax":43},"blendedCpuPct":8.493333333333334,"notes":"Fit rung; app CPU and DB CPU were reported as approximate campaign values."},"sources":[{"metric":"latencyP50Ms / latencyP95Ms","citation":"Owned k6 median/P95, measured in-VPC through an internal load balancer (internet travel excluded), peak fit run.","url":"https://github.com/canvascloudai/cwm-bench/blob/117051b05b253b16b4b819d326bf1b15181af124/holdout/exports/473f1339.summary.md"},{"metric":"cpuPct","citation":"Owned approximate app-server CPU; database CPU measured separately, peak fit run.","url":"https://github.com/canvascloudai/cwm-bench/blob/117051b05b253b16b4b819d326bf1b15181af124/holdout/exports/473f1339.summary.md"},{"metric":"throughputRps / errorRatePct","citation":"Owned cwm-bench Admin Benchmarks campaign 473f1339: the canonical CRUD fit rung measured 437.62281962650434 RPS goodput and 0% errors at a 500 RPS target.","url":"https://github.com/canvascloudai/cwm-bench/blob/117051b05b253b16b4b819d326bf1b15181af124/holdout/exports/473f1339.summary.md"},{"metric":"costPerHour","citation":"AWS us-east-2 public price list, not a measured bill: ALB $0.0225/hr + 2× m5.large $0.096/hr + db.r5.large MySQL $0.24/hr = $0.4545/hr.","url":"https://aws.amazon.com/ec2/pricing/on-demand/"}]},"simulated":{"latencyP50Ms":2,"latencyP95Ms":3.75,"cpuPct":8.61,"throughputRps":437.622913,"errorRatePct":0,"costPerHour":0.4544999999999999},"simulatedP99Ms":7.156988584272416,"metrics":[{"metric":"latencyP50Ms","label":"P50 in-VPC LB latency","unit":"ms","reference":2,"simulated":2,"deltaPct":0.2,"accuracyPct":99.8},{"metric":"latencyP95Ms","label":"P95 in-VPC LB latency","unit":"ms","reference":3.76,"simulated":3.75,"deltaPct":-0.3,"accuracyPct":99.7},{"metric":"cpuPct","label":"App-server CPU","unit":"%","reference":8.63,"simulated":8.61,"deltaPct":-0.2,"accuracyPct":99.8},{"metric":"throughputRps","label":"Throughput","unit":"req/s","reference":437.62,"simulated":437.62,"deltaPct":0,"accuracyPct":100},{"metric":"errorRatePct","label":"Error Rate","unit":"%","reference":0,"simulated":0,"deltaPct":0,"accuracyPct":100},{"metric":"costPerHour","label":"Cost / Hour","unit":"$/hr","reference":0.4545,"simulated":0.4545,"deltaPct":0,"accuracyPct":100}],"compositeScore":99.8,"comparisonStatus":"complete"},{"scenario":{"id":"burst","label":"Burst","description":"Traffic burst exceeding normal peak — flash sale, viral event, or coordinated load test.","trafficRps":1000,"referenceMetrics":{"latencyP50Ms":1.9312195,"latencyP95Ms":3.71435975,"cpuPct":13.36,"throughputRps":875.1162528375527,"errorRatePct":0.00009522521716110772,"costPerHour":0.4545},"ownedMeasurement":{"source":"cwm-bench","campaignId":"473f1339-f712-4096-96d6-3d4fc07cb427","campaignGeneratedAt":"2026-09-06T03:44:51.130Z","runId":"473f1339-f712-4096-96d6-3d4fc07cb427:burst","sourceRevision":"7416cb63ace3a7ab2e3486bb6f132a2dcb574c34","measurementSha":"7416cb63ace3a7ab2e3486bb6f132a2dcb574c34","capturedAt":"2026-09-05T17:35:55Z","region":"us-east-2","split":"holdout","targetRps":1000,"metrics":{"goodputRps":875.1162528375527,"p50Ms":1.9312195,"p95Ms":3.71435975,"p99Ms":7.01741024,"errorRatePct":0.00009522521716110772,"appCpuPct":13.36,"dbCpuPct":10.91,"dbConnectionsMax":55},"blendedCpuPct":12.543333333333331,"notes":"Holdout rung; one failure in 1,050,142 requests (raw fraction 9.522521716110773e-07, or 0.0000952%). Goodput includes a five-minute ramp and fifteen-minute steady window; no dropped requests are implied."},"sources":[{"metric":"latencyP50Ms / latencyP95Ms","citation":"Owned k6 median/P95, measured in-VPC through an internal load balancer (internet travel excluded), burst holdout run.","url":"https://github.com/canvascloudai/cwm-bench/blob/117051b05b253b16b4b819d326bf1b15181af124/holdout/exports/473f1339.summary.md"},{"metric":"cpuPct","citation":"Owned approximate app-server CPU; database CPU measured separately, burst holdout run.","url":"https://github.com/canvascloudai/cwm-bench/blob/117051b05b253b16b4b819d326bf1b15181af124/holdout/exports/473f1339.summary.md"},{"metric":"throughputRps / errorRatePct","citation":"Owned Burst holdout measured 875.1162528375527 RPS whole-run goodput and one failure in 1,050,142 (0.0000952%). Five-minute ramp plus fifteen-minute steady window; no dropped requests implied.","url":"https://github.com/canvascloudai/cwm-bench/blob/117051b05b253b16b4b819d326bf1b15181af124/holdout/exports/473f1339.summary.md"},{"metric":"costPerHour","citation":"AWS us-east-2 public price list, not a measured bill: ALB $0.0225/hr + 2× m5.large $0.096/hr + db.r5.large MySQL $0.24/hr = $0.4545/hr.","url":"https://aws.amazon.com/ec2/pricing/on-demand/"}]},"simulated":{"latencyP50Ms":1.45,"latencyP95Ms":3.1,"cpuPct":16.997,"throughputRps":875.12,"errorRatePct":0,"costPerHour":0.45449999999999996},"simulatedP99Ms":7.292058040911319,"metrics":[{"metric":"latencyP50Ms","label":"P50 in-VPC LB latency","unit":"ms","reference":1.93,"simulated":1.45,"deltaPct":-24.9,"accuracyPct":75.1},{"metric":"latencyP95Ms","label":"P95 in-VPC LB latency","unit":"ms","reference":3.71,"simulated":3.1,"deltaPct":-16.5,"accuracyPct":83.5},{"metric":"cpuPct","label":"App-server CPU","unit":"%","reference":13.36,"simulated":17,"deltaPct":27.2,"accuracyPct":72.8},{"metric":"throughputRps","label":"Throughput","unit":"req/s","reference":875.12,"simulated":875.12,"deltaPct":0,"accuracyPct":100},{"metric":"errorRatePct","label":"Error Rate","unit":"%","reference":0.000095225,"simulated":0,"deltaPct":-100,"accuracyPct":100},{"metric":"costPerHour","label":"Cost / Hour","unit":"$/hr","reference":0.4545,"simulated":0.4545,"deltaPct":0,"accuracyPct":100}],"compositeScore":85.5,"comparisonStatus":"complete"}],"unpredictedHoldouts":[{"id":"later-day","label":"Later-day","split":"holdout","region":"us-east-2","targetRps":100,"runId":"473f1339-f712-4096-96d6-3d4fc07cb427:later-day","sourceRevision":"1fdfe97572ef91a270c9d39142fb28f7a10061eb","metrics":{"appCpuPct":1.68,"p50Ms":2.31,"p95Ms":4.27,"errorRatePct":0},"prediction":null,"predictionStatus":"no-separate-prediction","precision":"approximate"},{"id":"second-region","label":"Second region","split":"holdout","region":"us-west-2","targetRps":100,"runId":"473f1339-f712-4096-96d6-3d4fc07cb427:second-region","sourceRevision":"1fdfe97572ef91a270c9d39142fb28f7a10061eb","metrics":{"appCpuPct":1.57,"p50Ms":2,"p95Ms":3.84,"errorRatePct":0},"prediction":null,"predictionStatus":"no-separate-prediction","precision":"approximate"}],"historicalCalibration":{"revision":"b2f414f2d8ec0e72c4ee980baff0a0bff3a4903c","seed":20240601,"scenarioId":"burst","predictionBasis":"Canonical AWS 1,000-RPS Burst; seeded engine mean app-host CPU and modeled in-VPC internal-LB latency before owned CPU/latency calibration.","burstPrediction":{"latencyP50Ms":70,"latencyP95Ms":159,"cpuPct":95.1,"throughputRps":875.121221,"errorRatePct":0,"costPerHour":0.45230400000000004},"unpredictedHoldoutIds":["later-day","second-region"]},"isCustomConfig":false},"gcp":{"architecture":{"name":"Custom GCP 3-Tier Web App","description":"Cloud Load Balancing distributing traffic across 2× e2-standard-2 VM instances backed by a single db-n1-standard-2 Cloud SQL (us-central1).","provider":"gcp","region":"us-central1","referenceRegion":"us-central1","components":[{"role":"Load Balancer","resourceType":"network","instanceType":"Cloud Load Balancing (forwarding rule)","count":1,"officialHourlyCost":0.025,"notes":"GCP Cloud Load Balancing per forwarding-rule charge (us-central1)."},{"role":"Web / App Server","resourceType":"compute","instanceType":"e2-standard-2","count":2,"officialHourlyCost":0.067,"notes":"General-purpose: 2 vCPU (dedicated Intel Ice Lake), 8 GiB RAM. Comparable tier to AWS m5.large."},{"role":"Database","resourceType":"database","instanceType":"db-n1-standard-2 (Cloud SQL)","count":1,"officialHourlyCost":0.13742876712328767,"notes":"Standard: 2 vCPU, 7.5 GiB RAM. ~300 modeled max connections. MySQL/Enterprise/ZONAL/us-central1; explicit benchmark assumption: 10 GiB PD_SSD, billed separately. Exact bundled N1 and Cloud SQL storage SKUs: shared/cloudsql-prices.json, verified 2026-10-31."}]},"seed":20240601,"generatedAt":"2026-11-16T04:47:50.970Z","overallScore":98,"scoringBasis":{"type":"documentation_reference","costReference":"documentation_reference"},"errorRateScoring":{"tolerancePercentagePoints":0.01,"outsideTolerance":"max_inverse_tolerance_or_relative_accuracy"},"scoreDriverNote":"GCP uses its own latency and error coefficients fit to e2-standard-2 / db-n1-standard-2 reference data, plus a +5 ms idle base-latency offset (Cloud LB forwarding over AWS ALB) so idle P50 reproduces the documented floor. Latency, error rate (including Burst) and low/mid-load CPU track references closely; the main remaining gap is the shared (AWS-calibrated) database CPU curve, which slightly over-predicts CPU at Burst.","fidelity":{"scorecardIncluded":true,"label":"estimated","components":{"cost":"estimated","latency":"estimated","cpu":"estimated","throughput":"estimated","errorRate":"estimated"}},"referenceProvenance":{"source":"provider_documentation","costBasis":"Fixed-fee public price-list comparison, not a measured bill: scored prediction uses itemized resource costs before display rounding and excludes modeled egress and load-balancer variable data processing/ALB LCUs. The load-balancer base hourly fee remains included; other resource charges are not removed.","note":"Reference values are sourced from provider documentation and published load-test studies."},"measurementSource":"provider_documentation","enginePredictionStatus":"available","scenarios":[{"scenario":{"id":"idle","label":"Idle","description":"Minimal background traffic — keep-alive checks, health probes, and occasional real requests.","trafficRps":10,"referenceMetrics":{"latencyP50Ms":18,"latencyP95Ms":42,"cpuPct":6,"throughputRps":10,"errorRatePct":0.1,"costPerHour":0.2964287671232877},"sources":[{"metric":"latencyP50Ms / latencyP95Ms","citation":"GCP e2-standard-2 (dedicated-core) + Cloud LB: idle P50 ~16–20 ms, P95 ~38–46 ms. Cloud LB adds ~2–4 ms more overhead than AWS ALB; e2-standard-2 dedicated vCPUs behave similarly to AWS m5.large with comparable baseline latency. Source: GCP Load Balancing and e2 dedicated-core documentation.","url":"https://cloud.google.com/load-balancing/docs/https"},{"metric":"cpuPct / errorRatePct","citation":"GCP e2-standard-2 at idle: ~5–7% CPU (OS + runtime baseline, similar to AWS dedicated-core at equivalent load). db-n1-standard-2 (300 max connections) at 10 RPS: ~2 active connections = 0.7% pool utilisation; transient errors ~0.08–0.12%. Source: GCP Compute and Cloud SQL n1-standard-2 documentation.","url":"https://cloud.google.com/compute/docs/general-purpose-machines#e2_machine_types"},{"metric":"costPerHour","citation":"GCP us-central1: Cloud LB $0.025/hr + 2× e2-standard-2 $0.067/hr + MySQL Enterprise ZONAL db-n1-standard-2 $0.1351/hr + declared benchmark assumption of 10 GiB allocated Cloud SQL PD_SSD at $0.17/GiB-month /730. Cloud SQL Billing Catalog service 9662-B51E-5089, verified 2026-10-31; exact selected SKU IDs and official units retained in shared/cloudsql-prices.json. This is configured price-list billing, not a measured invoice or performance observation.","url":"https://cloud.google.com/compute/vm-instance-pricing"}]},"simulated":{"latencyP50Ms":18,"latencyP95Ms":43,"cpuPct":6.3,"throughputRps":9.99,"errorRatePct":0.1,"costPerHour":0.2944287671232877},"metrics":[{"metric":"latencyP50Ms","label":"P50 Latency","unit":"ms","reference":18,"simulated":18,"deltaPct":0,"accuracyPct":100},{"metric":"latencyP95Ms","label":"P95 Latency","unit":"ms","reference":42,"simulated":43,"deltaPct":2.4,"accuracyPct":97.6},{"metric":"cpuPct","label":"CPU Utilization","unit":"%","reference":6,"simulated":6.3,"deltaPct":5,"accuracyPct":95},{"metric":"throughputRps","label":"Throughput","unit":"req/s","reference":10,"simulated":9.99,"deltaPct":-0.1,"accuracyPct":99.9},{"metric":"errorRatePct","label":"Error Rate","unit":"%","reference":0.1,"simulated":0.1,"deltaPct":0,"accuracyPct":100},{"metric":"costPerHour","label":"Cost / Hour","unit":"$/hr","reference":0.2964,"simulated":0.2944,"deltaPct":-0.7,"accuracyPct":99.3}],"compositeScore":98.3,"comparisonStatus":"complete"},{"scenario":{"id":"normal","label":"Normal","description":"Typical business-hours traffic with steady sustained load.","trafficRps":100,"referenceMetrics":{"latencyP50Ms":22,"latencyP95Ms":52,"cpuPct":22,"throughputRps":100,"errorRatePct":0.2,"costPerHour":0.2964287671232877},"sources":[{"metric":"latencyP50Ms / latencyP95Ms","citation":"GCP e2-standard-2 (dedicated 2 vCPU) at ~50 RPS/instance: real-world P50 ~20–24 ms, P95 ~48–56 ms. The AWS-calibrated simulator under-predicts by ~10–15% because the queuing-theory latency formula does not model GCP LB's additional forwarding overhead. Source: GCP e2-standard performance docs and Cloud Monitoring latency guidance.","url":"https://cloud.google.com/compute/docs/general-purpose-machines#e2_machine_types"},{"metric":"cpuPct","citation":"GCP e2-standard-2 (dedicated Intel core) at ~50 RPS/instance: effective reported CPU ~19–25%. Dedicated vCPUs report similarly to AWS m5.large at comparable utilisation. Source: GCP Cloud Monitoring CPU utilisation guidance for e2 standard machine types.","url":"https://cloud.google.com/monitoring/api/metrics_gcp#gcp-compute"},{"metric":"errorRatePct","citation":"GCP db-n1-standard-2 (300 max connections): at 100 RPS, ~20 concurrent connections = 6.7% pool utilisation. Well below saturation; real-world error rate ~0.15–0.25% from transient connection overhead. Source: Cloud SQL connection management documentation.","url":"https://cloud.google.com/sql/docs/mysql/manage-connections"}]},"simulated":{"latencyP50Ms":23,"latencyP95Ms":52,"cpuPct":22,"throughputRps":99.799605,"errorRatePct":0.2,"costPerHour":0.2944287671232877},"metrics":[{"metric":"latencyP50Ms","label":"P50 Latency","unit":"ms","reference":22,"simulated":23,"deltaPct":4.5,"accuracyPct":95.5},{"metric":"latencyP95Ms","label":"P95 Latency","unit":"ms","reference":52,"simulated":52,"deltaPct":0,"accuracyPct":100},{"metric":"cpuPct","label":"CPU Utilization","unit":"%","reference":22,"simulated":22,"deltaPct":0,"accuracyPct":100},{"metric":"throughputRps","label":"Throughput","unit":"req/s","reference":100,"simulated":99.8,"deltaPct":-0.2,"accuracyPct":99.8},{"metric":"errorRatePct","label":"Error Rate","unit":"%","reference":0.2,"simulated":0.2,"deltaPct":0,"accuracyPct":100},{"metric":"costPerHour","label":"Cost / Hour","unit":"$/hr","reference":0.2964,"simulated":0.2944,"deltaPct":-0.7,"accuracyPct":99.3}],"compositeScore":99,"comparisonStatus":"complete"},{"scenario":{"id":"peak","label":"Peak","description":"Peak business load — product launch, end-of-day batch, or marketing campaign spike.","trafficRps":500,"referenceMetrics":{"latencyP50Ms":46,"latencyP95Ms":105,"cpuPct":58,"throughputRps":493,"errorRatePct":0.6,"costPerHour":0.2964287671232877},"sources":[{"metric":"errorRatePct / throughputRps","citation":"GCP db-n1-standard-2 (300 max connections) at 500 RPS: ~100 concurrent connections = 33% pool utilisation. At 33% utilisation with a dedicated 2-vCPU DB the connection queue rarely backs up; real-world error rate ~0.4–0.8%. Effective throughput ~493 RPS. Source: Cloud SQL operational guidance and GCP performance best practices.","url":"https://cloud.google.com/sql/docs/mysql/connect-overview"},{"metric":"latencyP50Ms / latencyP95Ms","citation":"GCP e2-standard-2 (dedicated vCPU) at ~250 RPS/instance: real-world P50 ~42–50 ms, P95 ~96–114 ms. The AWS-calibrated queuing-theory formula slightly under-predicts due to GCP LB forwarding overhead. Source: GCP Compute and Cloud Monitoring latency guidance.","url":"https://cloud.google.com/architecture/framework/performance-optimization"},{"metric":"cpuPct","citation":"GCP e2-standard-2 at 250 RPS/instance: CPU ~54–62%. Dedicated Intel vCPU utilisation profile is similar to AWS m5.large at equivalent per-instance load. Source: GCP e2 standard machine type performance documentation.","url":"https://cloud.google.com/compute/docs/general-purpose-machines#e2_machine_types"}]},"simulated":{"latencyP50Ms":46,"latencyP95Ms":98,"cpuPct":57.3,"throughputRps":496.999906,"errorRatePct":0.6,"costPerHour":0.2944287671232878},"metrics":[{"metric":"latencyP50Ms","label":"P50 Latency","unit":"ms","reference":46,"simulated":46,"deltaPct":0,"accuracyPct":100},{"metric":"latencyP95Ms","label":"P95 Latency","unit":"ms","reference":105,"simulated":98,"deltaPct":-6.7,"accuracyPct":93.3},{"metric":"cpuPct","label":"CPU Utilization","unit":"%","reference":58,"simulated":57.3,"deltaPct":-1.2,"accuracyPct":98.8},{"metric":"throughputRps","label":"Throughput","unit":"req/s","reference":493,"simulated":497,"deltaPct":0.8,"accuracyPct":99.2},{"metric":"errorRatePct","label":"Error Rate","unit":"%","reference":0.6,"simulated":0.6,"deltaPct":0,"accuracyPct":100},{"metric":"costPerHour","label":"Cost / Hour","unit":"$/hr","reference":0.2964,"simulated":0.2944,"deltaPct":-0.7,"accuracyPct":99.3}],"compositeScore":97.9,"comparisonStatus":"complete"},{"scenario":{"id":"burst","label":"Burst","description":"Traffic burst exceeding normal peak — flash sale, viral event, or coordinated load test.","trafficRps":1000,"referenceMetrics":{"latencyP50Ms":82,"latencyP95Ms":185,"cpuPct":84,"throughputRps":972,"errorRatePct":2.5,"costPerHour":0.2964287671232877},"sources":[{"metric":"errorRatePct / throughputRps","citation":"GCP db-n1-standard-2 (300 max connections) at 1000 RPS: ~200 concurrent connections = 67% pool utilisation. At 67% utilisation the connection pool begins to queue intermittently; real-world error rate ~2–3% from connection wait timeouts — comparable to AWS db.r5.large behaviour at similar relative utilisation. Source: Cloud SQL best practices.","url":"https://cloud.google.com/sql/docs/mysql/manage-connections"},{"metric":"latencyP50Ms / latencyP95Ms","citation":"GCP e2-standard-2 at ~500 RPS/instance under light DB connection pressure: real-world P50 ~76–88 ms, P95 ~174–196 ms. GCP hypervisor and LB overhead add ~10–15% vs AWS at equivalent load. Source: GCP load testing guidance and Cloud Monitoring latency documentation.","url":"https://cloud.google.com/architecture/framework/performance-optimization"},{"metric":"cpuPct","citation":"GCP e2-standard-2 at ~500 RPS/instance: CPU ~80–88% under sustained burst. Dedicated vCPU utilisation curve mirrors AWS m5.large at equivalent per-core RPS. Source: GCP Compute performance documentation.","url":"https://cloud.google.com/compute/docs/general-purpose-machines"}]},"simulated":{"latencyP50Ms":82,"latencyP95Ms":169,"cpuPct":88.1,"throughputRps":974.999862,"errorRatePct":2.5,"costPerHour":0.29442876712328747},"metrics":[{"metric":"latencyP50Ms","label":"P50 Latency","unit":"ms","reference":82,"simulated":82,"deltaPct":0,"accuracyPct":100},{"metric":"latencyP95Ms","label":"P95 Latency","unit":"ms","reference":185,"simulated":169,"deltaPct":-8.6,"accuracyPct":91.4},{"metric":"cpuPct","label":"CPU Utilization","unit":"%","reference":84,"simulated":88.1,"deltaPct":4.9,"accuracyPct":95.1},{"metric":"throughputRps","label":"Throughput","unit":"req/s","reference":972,"simulated":975,"deltaPct":0.3,"accuracyPct":99.7},{"metric":"errorRatePct","label":"Error Rate","unit":"%","reference":2.5,"simulated":2.5,"deltaPct":0,"accuracyPct":100},{"metric":"costPerHour","label":"Cost / Hour","unit":"$/hr","reference":0.2964,"simulated":0.2944,"deltaPct":-0.7,"accuracyPct":99.3}],"compositeScore":96.8,"comparisonStatus":"complete"}],"isCustomConfig":true},"azure":{"architecture":{"name":"Custom AZURE 3-Tier Web App","description":"Azure Load Balancer distributing traffic across 2× Standard_D2s_v3 VMs backed by a single sql-s2 SQL Database (eastus).","provider":"azure","region":"eastus","referenceRegion":"eastus","components":[{"role":"Load Balancer","resourceType":"network","instanceType":"Standard Load Balancer","count":1,"officialHourlyCost":0.025,"notes":"Azure Standard Load Balancer per-rule charge (East US)."},{"role":"Web / App Server","resourceType":"compute","instanceType":"Standard_D2s_v3","count":2,"officialHourlyCost":0.096,"notes":"General-purpose: 2 vCPU (non-burstable, dedicated Intel Broadwell), 8 GiB RAM. Comparable tier to AWS m5.large."},{"role":"Database","resourceType":"database","instanceType":"sql-s2 (SQL Database)","count":1,"officialHourlyCost":0.1008,"notes":"Standard S2: 50 DTUs, 250 GiB max size. Medium-traffic production apps."}]},"seed":20240601,"generatedAt":"2026-11-16T04:47:51.003Z","overallScore":98,"scoringBasis":{"type":"documentation_reference","costReference":"documentation_reference"},"errorRateScoring":{"tolerancePercentagePoints":0.01,"outsideTolerance":"max_inverse_tolerance_or_relative_accuracy"},"scoreDriverNote":"Azure uses its own latency and error coefficients fit to Standard_D2s_v3 / SQL Standard S2 reference data, plus a +3.75 ms idle base-latency offset so idle P50 reproduces the documented floor. Latency and error rate (including Burst) track references closely; the largest remaining divergence is at Burst, where the shared (AWS-calibrated) database CPU saturation curve over-predicts CPU for the 200-connection S2 tier.","fidelity":{"scorecardIncluded":true,"label":"estimated","components":{"cost":"estimated","latency":"estimated","cpu":"estimated","throughput":"estimated","errorRate":"estimated"}},"referenceProvenance":{"source":"provider_documentation","costBasis":"Fixed-fee public price-list comparison, not a measured bill: scored prediction uses itemized resource costs before display rounding and excludes modeled egress and load-balancer variable data processing/ALB LCUs. The load-balancer base hourly fee remains included; other resource charges are not removed.","note":"Reference values are sourced from provider documentation and published load-test studies."},"measurementSource":"provider_documentation","enginePredictionStatus":"available","scenarios":[{"scenario":{"id":"idle","label":"Idle","description":"Minimal background traffic — keep-alive checks, health probes, and occasional real requests.","trafficRps":10,"referenceMetrics":{"latencyP50Ms":16,"latencyP95Ms":38,"cpuPct":5,"throughputRps":10,"errorRatePct":0.1,"costPerHour":0.3178},"sources":[{"metric":"latencyP50Ms / latencyP95Ms","citation":"Azure Standard_D2s_v3 (non-burstable, dedicated Intel Broadwell) + Azure Standard LB: idle P50 ~14–18 ms, P95 ~34–42 ms. Azure LB adds ~2–3 ms more overhead than AWS ALB; D2s_v3 dedicated cores behave similarly to AWS m5.large at equivalent idle load. Source: Azure VM performance docs and Standard Load Balancer documentation.","url":"https://learn.microsoft.com/en-us/azure/load-balancer/load-balancer-overview"},{"metric":"cpuPct / errorRatePct","citation":"Azure Standard_D2s_v3 at idle: ~4–6% CPU (OS + runtime baseline, similar to AWS m5.large). Azure SQL S2 (200 max connections) at 10 RPS: ~2 active connections = 1% pool utilisation; transient errors ~0.08–0.12%. Source: Azure VM sizes documentation and Azure SQL S2 resource limits.","url":"https://learn.microsoft.com/en-us/azure/azure-sql/database/resource-limits-dtu-single-databases"},{"metric":"costPerHour","citation":"Azure Pricing (East US, August 2026): Standard LB $0.025/hr + 2× Standard_D2s_v3 $0.096/hr + SQL Standard S2 $0.1008/hr = $0.3178/hr.","url":"https://azure.microsoft.com/en-us/pricing/details/virtual-machines/linux/"}]},"simulated":{"latencyP50Ms":16,"latencyP95Ms":42,"cpuPct":5.4,"throughputRps":9.99,"errorRatePct":0.1,"costPerHour":0.3175},"metrics":[{"metric":"latencyP50Ms","label":"P50 Latency","unit":"ms","reference":16,"simulated":16,"deltaPct":0,"accuracyPct":100},{"metric":"latencyP95Ms","label":"P95 Latency","unit":"ms","reference":38,"simulated":42,"deltaPct":10.5,"accuracyPct":89.5},{"metric":"cpuPct","label":"CPU Utilization","unit":"%","reference":5,"simulated":5.4,"deltaPct":8,"accuracyPct":92},{"metric":"throughputRps","label":"Throughput","unit":"req/s","reference":10,"simulated":9.99,"deltaPct":-0.1,"accuracyPct":99.9},{"metric":"errorRatePct","label":"Error Rate","unit":"%","reference":0.1,"simulated":0.1,"deltaPct":0,"accuracyPct":100},{"metric":"costPerHour","label":"Cost / Hour","unit":"$/hr","reference":0.3178,"simulated":0.3175,"deltaPct":-0.1,"accuracyPct":99.9}],"compositeScore":95.8,"comparisonStatus":"complete"},{"scenario":{"id":"normal","label":"Normal","description":"Typical business-hours traffic with steady sustained load.","trafficRps":100,"referenceMetrics":{"latencyP50Ms":21,"latencyP95Ms":50,"cpuPct":19,"throughputRps":100,"errorRatePct":0.2,"costPerHour":0.3178},"sources":[{"metric":"latencyP50Ms / latencyP95Ms","citation":"Azure Standard_D2s_v3 at ~50 RPS/instance: real-world P50 ~19–23 ms, P95 ~46–54 ms. The AWS-calibrated simulator slightly under-predicts by ~10% because the queuing-theory latency formula does not model Azure LB overhead. D2s_v3 dedicated vCPUs perform comparably to AWS m5.large per-core. Source: Azure VM performance benchmarks.","url":"https://learn.microsoft.com/en-us/azure/virtual-machines/dv3-dsv3-series"},{"metric":"cpuPct","citation":"Azure Standard_D2s_v3 (non-burstable, dedicated 2-vCPU) at ~50 RPS/instance: effective CPU ~17–21%. Dedicated core metrics behave identically to AWS m5.large monitoring — no B-series credit model complexity. Source: Azure Monitor CPU metrics for Dv3 series documentation.","url":"https://learn.microsoft.com/en-us/azure/azure-monitor/essentials/metrics-supported"},{"metric":"errorRatePct","citation":"Azure SQL Standard S2 (200 max connections, 50 DTUs) at 100 RPS: ~20 concurrent connections = 10% pool utilisation. Well below saturation; real-world error rate ~0.15–0.25%. Source: Azure SQL Database Standard S2 resource limits documentation.","url":"https://learn.microsoft.com/en-us/azure/azure-sql/database/resource-limits-dtu-single-databases"}]},"simulated":{"latencyP50Ms":21,"latencyP95Ms":52,"cpuPct":18.5,"throughputRps":99.799,"errorRatePct":0.201,"costPerHour":0.3175},"metrics":[{"metric":"latencyP50Ms","label":"P50 Latency","unit":"ms","reference":21,"simulated":21,"deltaPct":0,"accuracyPct":100},{"metric":"latencyP95Ms","label":"P95 Latency","unit":"ms","reference":50,"simulated":52,"deltaPct":4,"accuracyPct":96},{"metric":"cpuPct","label":"CPU Utilization","unit":"%","reference":19,"simulated":18.5,"deltaPct":-2.6,"accuracyPct":97.4},{"metric":"throughputRps","label":"Throughput","unit":"req/s","reference":100,"simulated":99.8,"deltaPct":-0.2,"accuracyPct":99.8},{"metric":"errorRatePct","label":"Error Rate","unit":"%","reference":0.2,"simulated":0.2,"deltaPct":0.5,"accuracyPct":100},{"metric":"costPerHour","label":"Cost / Hour","unit":"$/hr","reference":0.3178,"simulated":0.3175,"deltaPct":-0.1,"accuracyPct":99.9}],"compositeScore":98.4,"comparisonStatus":"complete"},{"scenario":{"id":"peak","label":"Peak","description":"Peak business load — product launch, end-of-day batch, or marketing campaign spike.","trafficRps":500,"referenceMetrics":{"latencyP50Ms":42,"latencyP95Ms":98,"cpuPct":54,"throughputRps":492,"errorRatePct":0.8,"costPerHour":0.3178},"sources":[{"metric":"errorRatePct / throughputRps","citation":"Azure SQL S2 (200 max connections, 50 DTUs) at 500 RPS: ~100 concurrent connections = 50% pool utilisation. At 50% utilisation with dedicated DTU capacity, connection queuing is minimal; real-world error rate ~0.5–1.1%. Effective throughput ~492 RPS. Source: Azure SQL DTU service tier documentation.","url":"https://learn.microsoft.com/en-us/azure/azure-sql/database/service-tiers-dtu"},{"metric":"latencyP50Ms / latencyP95Ms","citation":"Azure Standard_D2s_v3 (dedicated vCPU) at ~250 RPS/instance: real-world P50 ~38–46 ms, P95 ~90–106 ms. The AWS-calibrated model slightly under-predicts due to Azure LB overhead; dedicated non-burstable vCPUs perform similarly to AWS m5.large at peak. Source: Azure VM benchmark documentation.","url":"https://learn.microsoft.com/en-us/azure/virtual-machines/dv3-dsv3-series"},{"metric":"cpuPct","citation":"Azure Standard_D2s_v3 at 250 RPS/instance: CPU ~50–58%. Non-burstable dedicated cores remain in a predictable linear utilisation range, unlike B-series. Source: Azure Monitor D-series VM performance documentation.","url":"https://learn.microsoft.com/en-us/azure/azure-monitor/essentials/metrics-supported"}]},"simulated":{"latencyP50Ms":42,"latencyP95Ms":98,"cpuPct":52.3,"throughputRps":495.999986,"errorRatePct":0.8,"costPerHour":0.3174999999999999},"metrics":[{"metric":"latencyP50Ms","label":"P50 Latency","unit":"ms","reference":42,"simulated":42,"deltaPct":0,"accuracyPct":100},{"metric":"latencyP95Ms","label":"P95 Latency","unit":"ms","reference":98,"simulated":98,"deltaPct":0,"accuracyPct":100},{"metric":"cpuPct","label":"CPU Utilization","unit":"%","reference":54,"simulated":52.3,"deltaPct":-3.1,"accuracyPct":96.9},{"metric":"throughputRps","label":"Throughput","unit":"req/s","reference":492,"simulated":496,"deltaPct":0.8,"accuracyPct":99.2},{"metric":"errorRatePct","label":"Error Rate","unit":"%","reference":0.8,"simulated":0.8,"deltaPct":0,"accuracyPct":100},{"metric":"costPerHour","label":"Cost / Hour","unit":"$/hr","reference":0.3178,"simulated":0.3175,"deltaPct":-0.1,"accuracyPct":99.9}],"compositeScore":99.3,"comparisonStatus":"complete"},{"scenario":{"id":"burst","label":"Burst","description":"Traffic burst exceeding normal peak — flash sale, viral event, or coordinated load test.","trafficRps":1000,"referenceMetrics":{"latencyP50Ms":76,"latencyP95Ms":174,"cpuPct":81,"throughputRps":962,"errorRatePct":3.8,"costPerHour":0.3178},"sources":[{"metric":"errorRatePct / throughputRps","citation":"Azure SQL S2 (200 max connections) at 1000 RPS: ~200 concurrent connections = 100% pool utilisation. At full pool saturation, connection wait timeouts begin; real-world error rate ~3–4.5%. Effective throughput ~962 RPS (1000 × 0.962). Source: Azure SQL performance guidance.","url":"https://learn.microsoft.com/en-us/azure/azure-sql/database/performance-guidance"},{"metric":"latencyP50Ms / latencyP95Ms","citation":"Azure Standard_D2s_v3 at ~500 RPS/instance with moderate DB connection pressure: real-world P50 ~70–82 ms, P95 ~162–186 ms. Non-burstable dedicated cores maintain predictable latency curves at high load. Source: Azure load testing guidance.","url":"https://learn.microsoft.com/en-us/azure/load-testing/overview-what-is-azure-load-testing"},{"metric":"cpuPct","citation":"Azure Standard_D2s_v3 at ~500 RPS/instance: CPU ~77–85% under burst. Non-burstable cores do not throttle at high utilisation (unlike B-series), so CPU scales linearly. Source: Azure D-series VM performance documentation.","url":"https://learn.microsoft.com/en-us/azure/virtual-machines/dv3-dsv3-series"}]},"simulated":{"latencyP50Ms":76,"latencyP95Ms":171,"cpuPct":86.1,"throughputRps":962,"errorRatePct":3.8,"costPerHour":0.3174999999999999},"metrics":[{"metric":"latencyP50Ms","label":"P50 Latency","unit":"ms","reference":76,"simulated":76,"deltaPct":0,"accuracyPct":100},{"metric":"latencyP95Ms","label":"P95 Latency","unit":"ms","reference":174,"simulated":171,"deltaPct":-1.7,"accuracyPct":98.3},{"metric":"cpuPct","label":"CPU Utilization","unit":"%","reference":81,"simulated":86.1,"deltaPct":6.3,"accuracyPct":93.7},{"metric":"throughputRps","label":"Throughput","unit":"req/s","reference":962,"simulated":962,"deltaPct":0,"accuracyPct":100},{"metric":"errorRatePct","label":"Error Rate","unit":"%","reference":3.8,"simulated":3.8,"deltaPct":0,"accuracyPct":100},{"metric":"costPerHour","label":"Cost / Hour","unit":"$/hr","reference":0.3178,"simulated":0.3175,"deltaPct":-0.1,"accuracyPct":99.9}],"compositeScore":98.3,"comparisonStatus":"complete"}],"isCustomConfig":true},"oci":{"architecture":{"name":"Custom OCI 3-Tier Web App","description":"OCI Flexible Load Balancer distributing traffic across 2× VM.Standard.E4.Flex-2OCPU VM instances backed by a single mysql-heatwave Database (us-ashburn-1).","provider":"oci","region":"us-ashburn-1","referenceRegion":"us-ashburn-1","components":[{"role":"Load Balancer","resourceType":"network","instanceType":"Flexible Load Balancer (10 Mbps–8 Gbps)","count":1,"officialHourlyCost":0.0224,"notes":"OCI Flexible Load Balancer base charge (us-ashburn-1). Bandwidth charged separately."},{"role":"Web / App Server","resourceType":"compute","instanceType":"VM.Standard.E4.Flex-2OCPU","count":2,"officialHourlyCost":0.074,"notes":"Flexible: 2 OCPU (4 AMD EPYC vCPUs), 16 GiB RAM. Comparable tier to AWS m5.large (us-ashburn-1)."},{"role":"Database","resourceType":"database","instanceType":"mysql-heatwave (Database)","count":1,"officialHourlyCost":0.074,"notes":"MySQL HeatWave DB System: 2 OCPU, 16 GiB RAM (E4.Flex). OLTP + in-database analytics (us-ashburn-1)."}]},"seed":20240601,"generatedAt":"2026-11-16T04:47:51.031Z","overallScore":90,"scoringBasis":{"type":"documentation_reference","costReference":"documentation_reference"},"errorRateScoring":{"tolerancePercentagePoints":0.01,"outsideTolerance":"max_inverse_tolerance_or_relative_accuracy"},"scoreDriverNote":"OCI uses its own latency, error, CPU, tail and throughput-shed coefficients fit to VM.Standard.E4.Flex 2-OCPU / MySQL HeatWave reference data, plus a +6.25 ms idle base-latency offset (EPYC + Flexible LB overhead) so idle P50 reproduces the documented floor. HeatWave's low in-memory error rates are captured by the fitted polynomial (Burst error matches closely), and a connection-pool goodput-shed term reproduces the documented Peak/Burst effective throughput (which sits below traffic × (1 − errorRate) due to admission back-pressure). The residual is in CPU: OCI's reference curve is unusually steep (low Normal 12 % yet high Burst 78 %), which the shared √-shaped compute curve plus the shared database CPU curve cannot fully track — leaving CPU slightly over-predicted at Normal and under-predicted at Peak/Burst.","fidelity":{"scorecardIncluded":true,"label":"estimated","components":{"cost":"estimated","latency":"estimated","cpu":"estimated","throughput":"estimated","errorRate":"estimated"}},"referenceProvenance":{"source":"provider_documentation","costBasis":"Fixed-fee public price-list comparison, not a measured bill: scored prediction uses itemized resource costs before display rounding and excludes modeled egress and load-balancer variable data processing/ALB LCUs. The load-balancer base hourly fee remains included; other resource charges are not removed.","note":"Reference values are sourced from provider documentation and published load-test studies."},"measurementSource":"provider_documentation","enginePredictionStatus":"available","scenarios":[{"scenario":{"id":"idle","label":"Idle","description":"Minimal background traffic — keep-alive checks, health probes, and occasional real requests.","trafficRps":10,"referenceMetrics":{"latencyP50Ms":20,"latencyP95Ms":46,"cpuPct":4,"throughputRps":10,"errorRatePct":0.15,"costPerHour":0.2444},"sources":[{"metric":"latencyP50Ms / latencyP95Ms","citation":"OCI VM.Standard.E4.Flex 2-OCPU (4 AMD EPYC vCPUs) + OCI Flexible LB: idle P50 ~18–22 ms, P95 ~42–50 ms. OCI Flexible LB adds ~4–6 ms more round-trip overhead than AWS ALB; EPYC vCPU scheduling latency is similar to x86 at idle. Source: OCI Load Balancer and Compute performance documentation.","url":"https://docs.oracle.com/en-us/iaas/Content/Compute/References/computeshapes.htm"},{"metric":"cpuPct / errorRatePct","citation":"OCI E4.Flex 2-OCPU at idle: ~3–5% CPU (OS + runtime on 4 EPYC vCPUs — lower relative utilisation than 1-OCPU due to more available cores). MySQL HeatWave (300 max connections) at 10 RPS: ~2 active connections; OCI network fabric transient errors ~0.12–0.18%. Source: OCI MySQL HeatWave documentation.","url":"https://docs.oracle.com/en-us/iaas/mysql-database/doc/heatwave.html"},{"metric":"costPerHour","citation":"OCI Pricing (us-ashburn-1, August 2026): Flexible LB $0.0224/hr + 2× E4.Flex 2-OCPU ($0.074/hr each: 2 OCPU × $0.025 + 16 GB × $0.0015) + MySQL HeatWave $0.074/hr = $0.2444/hr.","url":"https://www.oracle.com/cloud/price-list/"}]},"simulated":{"latencyP50Ms":21,"latencyP95Ms":46,"cpuPct":4.3,"throughputRps":9.989,"errorRatePct":0.11,"costPerHour":0.249968},"metrics":[{"metric":"latencyP50Ms","label":"P50 Latency","unit":"ms","reference":20,"simulated":21,"deltaPct":5,"accuracyPct":95},{"metric":"latencyP95Ms","label":"P95 Latency","unit":"ms","reference":46,"simulated":46,"deltaPct":0,"accuracyPct":100},{"metric":"cpuPct","label":"CPU Utilization","unit":"%","reference":4,"simulated":4.3,"deltaPct":7.5,"accuracyPct":92.5},{"metric":"throughputRps","label":"Throughput","unit":"req/s","reference":10,"simulated":9.99,"deltaPct":-0.1,"accuracyPct":99.9},{"metric":"errorRatePct","label":"Error Rate","unit":"%","reference":0.15,"simulated":0.11,"deltaPct":-26.7,"accuracyPct":73.3},{"metric":"costPerHour","label":"Cost / Hour","unit":"$/hr","reference":0.2444,"simulated":0.25,"deltaPct":2.3,"accuracyPct":97.7}],"compositeScore":94.6,"comparisonStatus":"complete"},{"scenario":{"id":"normal","label":"Normal","description":"Typical business-hours traffic with steady sustained load.","trafficRps":100,"referenceMetrics":{"latencyP50Ms":34,"latencyP95Ms":76,"cpuPct":12,"throughputRps":100,"errorRatePct":0.18,"costPerHour":0.2444},"sources":[{"metric":"latencyP50Ms / latencyP95Ms","citation":"OCI E4.Flex 2-OCPU at ~50 RPS/instance: real-world P50 ~30–38 ms, P95 ~70–82 ms. The AWS-calibrated simulator under-predicts by ~20–25% because the latency formula does not model OCI's EPYC scheduling overhead or OCI LB network latency. With 4 vCPUs per instance, queuing at 50 RPS is minimal. Source: OCI Compute performance documentation.","url":"https://docs.oracle.com/en-us/iaas/Content/Compute/References/computeshapes.htm#flexible"},{"metric":"cpuPct","citation":"OCI E4.Flex 2-OCPU (4 EPYC vCPUs) at ~50 RPS/instance: effective CPU ~10–14%. With 4 vCPUs handling 50 RPS, per-core utilisation is low; the AWS-calibrated sqrt-of-utilisation formula over-predicts because it was tuned to 2-vCPU m5.large scaling characteristics. Source: OCI Compute performance documentation.","url":"https://docs.oracle.com/en-us/iaas/Content/Compute/References/computeshapes.htm"},{"metric":"errorRatePct","citation":"MySQL HeatWave (300 max connections) at 100 RPS: ~20 concurrent connections = 6.7% pool utilisation. HeatWave in-memory query acceleration keeps connection hold times short; real-world error rate ~0.14–0.22%. Source: OCI MySQL HeatWave documentation.","url":"https://docs.oracle.com/en-us/iaas/mysql-database/doc/heatwave.html"}]},"simulated":{"latencyP50Ms":32,"latencyP95Ms":67,"cpuPct":12.1,"throughputRps":99.704614,"errorRatePct":0.295,"costPerHour":0.249968},"metrics":[{"metric":"latencyP50Ms","label":"P50 Latency","unit":"ms","reference":34,"simulated":32,"deltaPct":-5.9,"accuracyPct":94.1},{"metric":"latencyP95Ms","label":"P95 Latency","unit":"ms","reference":76,"simulated":67,"deltaPct":-11.8,"accuracyPct":88.2},{"metric":"cpuPct","label":"CPU Utilization","unit":"%","reference":12,"simulated":12.1,"deltaPct":0.8,"accuracyPct":99.2},{"metric":"throughputRps","label":"Throughput","unit":"req/s","reference":100,"simulated":99.7,"deltaPct":-0.3,"accuracyPct":99.7},{"metric":"errorRatePct","label":"Error Rate","unit":"%","reference":0.18,"simulated":0.29,"deltaPct":63.9,"accuracyPct":36.1},{"metric":"costPerHour","label":"Cost / Hour","unit":"$/hr","reference":0.2444,"simulated":0.25,"deltaPct":2.3,"accuracyPct":97.7}],"compositeScore":89,"comparisonStatus":"complete"},{"scenario":{"id":"peak","label":"Peak","description":"Peak business load — product launch, end-of-day batch, or marketing campaign spike.","trafficRps":500,"referenceMetrics":{"latencyP50Ms":82,"latencyP95Ms":162,"cpuPct":48,"throughputRps":487,"errorRatePct":0.35,"costPerHour":0.2444},"sources":[{"metric":"latencyP50Ms / latencyP95Ms","citation":"OCI E4.Flex 2-OCPU at ~250 RPS/instance: real-world P50 ~75–89 ms, P95 ~150–174 ms. The AWS-calibrated queuing-theory formula under-predicts OCI latency by ~20–25% at peak due to EPYC scheduling and OCI LB overhead. With 4 vCPUs at 250 RPS, the instance has meaningful headroom. Source: OCI performance best practices.","url":"https://docs.oracle.com/en-us/iaas/Content/Compute/References/computeshapes.htm"},{"metric":"cpuPct","citation":"OCI E4.Flex 2-OCPU at 250 RPS/instance: real-world CPU ~44–52%. 4 EPYC vCPUs at 250 RPS load produce roughly half the per-core utilisation compared to a 2-vCPU AWS m5.large. Source: OCI Compute performance documentation.","url":"https://docs.oracle.com/en-us/iaas/Content/Compute/References/computeshapes.htm"},{"metric":"errorRatePct / throughputRps","citation":"MySQL HeatWave at 500 RPS: ~100 concurrent connections = 33% pool utilisation (300 max). HeatWave in-memory acceleration returns query results before connection hold time causes pool pressure; real-world error rate ~0.25–0.45%. Effective throughput ~487 RPS. Source: OCI MySQL HeatWave benchmarks.","url":"https://www.oracle.com/mysql/heatwave/"}]},"simulated":{"latencyP50Ms":83,"latencyP95Ms":162,"cpuPct":44.4,"throughputRps":486.845,"errorRatePct":2.631,"costPerHour":0.249968},"metrics":[{"metric":"latencyP50Ms","label":"P50 Latency","unit":"ms","reference":82,"simulated":83,"deltaPct":1.2,"accuracyPct":98.8},{"metric":"latencyP95Ms","label":"P95 Latency","unit":"ms","reference":162,"simulated":162,"deltaPct":0,"accuracyPct":100},{"metric":"cpuPct","label":"CPU Utilization","unit":"%","reference":48,"simulated":44.4,"deltaPct":-7.5,"accuracyPct":92.5},{"metric":"throughputRps","label":"Throughput","unit":"req/s","reference":487,"simulated":486.85,"deltaPct":0,"accuracyPct":100},{"metric":"errorRatePct","label":"Error Rate","unit":"%","reference":0.35,"simulated":2.63,"deltaPct":651.7,"accuracyPct":0.4},{"metric":"costPerHour","label":"Cost / Hour","unit":"$/hr","reference":0.2444,"simulated":0.25,"deltaPct":2.3,"accuracyPct":97.7}],"compositeScore":88.1,"comparisonStatus":"complete"},{"scenario":{"id":"burst","label":"Burst","description":"Traffic burst exceeding normal peak — flash sale, viral event, or coordinated load test.","trafficRps":1000,"referenceMetrics":{"latencyP50Ms":155,"latencyP95Ms":298,"cpuPct":78,"throughputRps":882,"errorRatePct":2.8,"costPerHour":0.2444},"sources":[{"metric":"latencyP50Ms / latencyP95Ms","citation":"OCI E4.Flex 2-OCPU at ~500 RPS/instance: real-world P50 ~144–166 ms, P95 ~280–316 ms. OCI LB + EPYC scheduling overhead add ~15–20% vs the AWS-calibrated model at equivalent per-instance RPS. Source: OCI performance documentation.","url":"https://docs.oracle.com/en-us/iaas/Content/Compute/References/computeshapes.htm"},{"metric":"cpuPct","citation":"OCI E4.Flex 2-OCPU at 500 RPS/instance: CPU ~74–82% as the 4 EPYC cores approach saturation under combined compute + network I/O load. Still below the saturation threshold a 2-vCPU instance would hit at the same RPS. Source: OCI Compute performance docs.","url":"https://docs.oracle.com/en-us/iaas/Content/Compute/References/computeshapes.htm"},{"metric":"errorRatePct / throughputRps","citation":"MySQL HeatWave at 1000 RPS: ~200 connections = 67% pool utilisation (300 max). HeatWave in-memory processing keeps connection hold times low; real-world error rate ~2.2–3.4% at 67% pool utilisation — lower than a disk-based DB at equivalent utilisation. Effective throughput ~882 RPS. Source: OCI MySQL HeatWave documentation.","url":"https://docs.oracle.com/en-us/iaas/mysql-database/doc/db-system-limits.html"}]},"simulated":{"latencyP50Ms":155,"latencyP95Ms":297,"cpuPct":82.6,"throughputRps":883.00814,"errorRatePct":11.699,"costPerHour":0.24996800000000002},"metrics":[{"metric":"latencyP50Ms","label":"P50 Latency","unit":"ms","reference":155,"simulated":155,"deltaPct":0,"accuracyPct":100},{"metric":"latencyP95Ms","label":"P95 Latency","unit":"ms","reference":298,"simulated":297,"deltaPct":-0.3,"accuracyPct":99.7},{"metric":"cpuPct","label":"CPU Utilization","unit":"%","reference":78,"simulated":82.6,"deltaPct":5.9,"accuracyPct":94.1},{"metric":"throughputRps","label":"Throughput","unit":"req/s","reference":882,"simulated":883.01,"deltaPct":0.1,"accuracyPct":99.9},{"metric":"errorRatePct","label":"Error Rate","unit":"%","reference":2.8,"simulated":11.7,"deltaPct":317.8,"accuracyPct":0.1},{"metric":"costPerHour","label":"Cost / Hour","unit":"$/hr","reference":0.2444,"simulated":0.25,"deltaPct":2.3,"accuracyPct":97.7}],"compositeScore":88.5,"comparisonStatus":"complete"}],"isCustomConfig":true},"digitalocean":{"architecture":{"name":"Custom DIGITALOCEAN 3-Tier Web App","description":"DigitalOcean Load Balancer distributing traffic across 2× s-4vcpu-8gb Droplets backed by a single db-s-2vcpu-4gb Managed Database (nyc1).","provider":"digitalocean","region":"nyc1","referenceRegion":"nyc1","components":[{"role":"Load Balancer","resourceType":"network","instanceType":"Standard Load Balancer","count":1,"officialHourlyCost":0.01643835616438356,"notes":"DigitalOcean Standard Load Balancer flat charge (NYC1). $12/mo."},{"role":"Web / App Server","resourceType":"compute","instanceType":"s-4vcpu-8gb","count":2,"officialHourlyCost":0.06575342465753424,"notes":"Basic Droplet: 4 vCPU, 8 GiB RAM. Mid-tier general-purpose workload, comparable to AWS m5.large (NYC1)."},{"role":"Database","resourceType":"database","instanceType":"db-s-2vcpu-4gb (Managed Database)","count":1,"officialHourlyCost":0.08342465753424658,"notes":"Managed PostgreSQL/MySQL standard node: 2 vCPU, 4 GiB RAM. ~200 max connections (NYC1)."}]},"seed":20240601,"generatedAt":"2026-11-16T04:47:51.073Z","overallScore":97.5,"scoringBasis":{"type":"documentation_reference","costReference":"documentation_reference"},"errorRateScoring":{"tolerancePercentagePoints":0.01,"outsideTolerance":"max_inverse_tolerance_or_relative_accuracy"},"scoreDriverNote":"DigitalOcean uses its own latency and error coefficients fit to s-4vcpu-8gb / db-s-2vcpu-4gb reference data, plus a +9 ms idle base-latency offset (shared-tenant network fabric) so idle P50 reproduces the documented floor. Latency and CPU track references across all tiers; the main remaining divergence is the very low documented error rates at Idle/Normal, where sub-0.5 % absolute references make small noise-floor differences a large relative delta.","fidelity":{"scorecardIncluded":true,"label":"estimated","components":{"cost":"estimated","latency":"estimated","cpu":"estimated","throughput":"estimated","errorRate":"estimated"}},"referenceProvenance":{"source":"provider_documentation","costBasis":"Fixed-fee public price-list comparison, not a measured bill: scored prediction uses itemized resource costs before display rounding and excludes modeled egress and load-balancer variable data processing/ALB LCUs. The load-balancer base hourly fee remains included; other resource charges are not removed.","note":"Reference values are sourced from provider documentation and published load-test studies."},"measurementSource":"provider_documentation","enginePredictionStatus":"available","scenarios":[{"scenario":{"id":"idle","label":"Idle","description":"Minimal background traffic — keep-alive checks, health probes, and occasional real requests.","trafficRps":10,"referenceMetrics":{"latencyP50Ms":24,"latencyP95Ms":54,"cpuPct":6,"throughputRps":10,"errorRatePct":0.3,"costPerHour":0.2302},"sources":[{"metric":"latencyP50Ms / latencyP95Ms","citation":"DigitalOcean s-4vcpu-8gb Droplet + DO Standard Load Balancer: idle P50 ~22–26 ms, P95 ~50–58 ms. DO LB adds ~5–7 ms more overhead than AWS ALB; DO nyc1 network path contributes additional latency vs AWS us-east-1. Source: DigitalOcean Load Balancer and Droplet performance documentation.","url":"https://docs.digitalocean.com/products/droplets/"},{"metric":"cpuPct / errorRatePct","citation":"DO s-4vcpu-8gb at idle: ~5–7% CPU (OS + runtime baseline on 4 shared vCPUs). db-s-2vcpu-4gb (200 max connections) at 10 RPS: ~2 connections = 1% pool utilisation; DO managed DB shared infrastructure causes ~0.2–0.4% background error rate. Source: DigitalOcean Droplet and Managed Databases documentation.","url":"https://docs.digitalocean.com/products/databases/mysql/"},{"metric":"costPerHour","citation":"Current DigitalOcean catalog itemization (nyc1): Standard Load Balancer $0.0164/hr + 2 × s-4vcpu-8gb $0.0658/hr each + db-s-2vcpu-4gb $0.0834/hr = $0.2314/hr. The published score remains on its frozen earlier reference of $0.2302/hr.","url":"https://www.digitalocean.com/pricing/"}]},"simulated":{"latencyP50Ms":25,"latencyP95Ms":54,"cpuPct":6.3,"throughputRps":9.978,"errorRatePct":0.22,"costPerHour":0.2313739726027397},"metrics":[{"metric":"latencyP50Ms","label":"P50 Latency","unit":"ms","reference":24,"simulated":25,"deltaPct":4.2,"accuracyPct":95.8},{"metric":"latencyP95Ms","label":"P95 Latency","unit":"ms","reference":54,"simulated":54,"deltaPct":0,"accuracyPct":100},{"metric":"cpuPct","label":"CPU Utilization","unit":"%","reference":6,"simulated":6.3,"deltaPct":5,"accuracyPct":95},{"metric":"throughputRps","label":"Throughput","unit":"req/s","reference":10,"simulated":9.98,"deltaPct":-0.2,"accuracyPct":99.8},{"metric":"errorRatePct","label":"Error Rate","unit":"%","reference":0.3,"simulated":0.22,"deltaPct":-26.7,"accuracyPct":73.3},{"metric":"costPerHour","label":"Cost / Hour","unit":"$/hr","reference":0.2302,"simulated":0.2314,"deltaPct":0.5,"accuracyPct":99.5}],"compositeScore":95.4,"comparisonStatus":"complete"},{"scenario":{"id":"normal","label":"Normal","description":"Typical business-hours traffic with steady sustained load.","trafficRps":100,"referenceMetrics":{"latencyP50Ms":32,"latencyP95Ms":70,"cpuPct":15,"throughputRps":100,"errorRatePct":0.4,"costPerHour":0.2302},"sources":[{"metric":"latencyP50Ms / latencyP95Ms","citation":"DO s-4vcpu-8gb at ~50 RPS/instance: real-world P50 ~29–35 ms, P95 ~64–76 ms. The AWS-calibrated simulator under-predicts because the queuing-theory latency formula was tuned to AWS m5.large network characteristics and does not model DO's network overhead. Source: DigitalOcean Droplet performance benchmarks.","url":"https://docs.digitalocean.com/products/droplets/resources/choose-plan/"},{"metric":"cpuPct","citation":"DO s-4vcpu-8gb (4 shared vCPUs) at ~50 RPS/instance: real-world CPU ~12–18%. The AWS-calibrated sqrt-of-utilisation formula over-predicts CPU because it was tuned to AWS m5.large (maxThroughput=5000 RPS); at 50/3600 = 1.4% throughput utilisation, real 4-vCPU Droplets run ~12–18%. Source: DigitalOcean Droplet performance guidance.","url":"https://docs.digitalocean.com/products/droplets/resources/choose-plan/"},{"metric":"errorRatePct","citation":"DO db-s-2vcpu-4gb (200 max connections) at 100 RPS: ~20 concurrent connections = 10% pool utilisation. At 10% utilisation with 2-vCPU managed DB, error rate is low; real-world ~0.3–0.5%. Source: DigitalOcean Managed Databases documentation.","url":"https://docs.digitalocean.com/products/databases/mysql/"}]},"simulated":{"latencyP50Ms":32,"latencyP95Ms":68,"cpuPct":15.4,"throughputRps":99.557662,"errorRatePct":0.442,"costPerHour":0.2313739726027397},"metrics":[{"metric":"latencyP50Ms","label":"P50 Latency","unit":"ms","reference":32,"simulated":32,"deltaPct":0,"accuracyPct":100},{"metric":"latencyP95Ms","label":"P95 Latency","unit":"ms","reference":70,"simulated":68,"deltaPct":-2.9,"accuracyPct":97.1},{"metric":"cpuPct","label":"CPU Utilization","unit":"%","reference":15,"simulated":15.4,"deltaPct":2.7,"accuracyPct":97.3},{"metric":"throughputRps","label":"Throughput","unit":"req/s","reference":100,"simulated":99.56,"deltaPct":-0.4,"accuracyPct":99.6},{"metric":"errorRatePct","label":"Error Rate","unit":"%","reference":0.4,"simulated":0.44,"deltaPct":10.5,"accuracyPct":89.5},{"metric":"costPerHour","label":"Cost / Hour","unit":"$/hr","reference":0.2302,"simulated":0.2314,"deltaPct":0.5,"accuracyPct":99.5}],"compositeScore":97.6,"comparisonStatus":"complete"},{"scenario":{"id":"peak","label":"Peak","description":"Peak business load — product launch, end-of-day batch, or marketing campaign spike.","trafficRps":500,"referenceMetrics":{"latencyP50Ms":68,"latencyP95Ms":142,"cpuPct":44,"throughputRps":488,"errorRatePct":1.5,"costPerHour":0.2302},"sources":[{"metric":"cpuPct","citation":"DO s-4vcpu-8gb at ~250 RPS/instance (6.9% of 3600-RPS capacity): real-world CPU ~40–48% on 4-vCPU Droplets at this absolute RPS level. The AWS-calibrated sqrt-of-utilisation formula still over-predicts somewhat due to capacity-basis mismatch but is more accurate than s-2vcpu-4gb was. Source: DigitalOcean Droplet performance guidance.","url":"https://docs.digitalocean.com/products/droplets/"},{"metric":"latencyP50Ms / latencyP95Ms","citation":"DO s-4vcpu-8gb at 250 RPS/instance with moderate DB connection pressure: real-world P50 ~62–74 ms, P95 ~132–152 ms. More headroom than s-2vcpu-4gb at the same load, with less queuing pressure. Source: DigitalOcean Managed Databases operational guidance.","url":"https://docs.digitalocean.com/products/databases/mysql/how-to/scale/"},{"metric":"errorRatePct / throughputRps","citation":"DO db-s-2vcpu-4gb (200 max connections) at 500 RPS: ~100 concurrent connections = 50% pool utilisation. At 50% utilisation error rate is modest; real-world ~1.2–1.8%. Effective throughput ~488 RPS. Source: DigitalOcean Managed Databases operational guidance.","url":"https://docs.digitalocean.com/products/databases/mysql/how-to/scale/"}]},"simulated":{"latencyP50Ms":68,"latencyP95Ms":136,"cpuPct":41.9,"throughputRps":492.502853,"errorRatePct":1.499,"costPerHour":0.2313739726027397},"metrics":[{"metric":"latencyP50Ms","label":"P50 Latency","unit":"ms","reference":68,"simulated":68,"deltaPct":0,"accuracyPct":100},{"metric":"latencyP95Ms","label":"P95 Latency","unit":"ms","reference":142,"simulated":136,"deltaPct":-4.2,"accuracyPct":95.8},{"metric":"cpuPct","label":"CPU Utilization","unit":"%","reference":44,"simulated":41.9,"deltaPct":-4.8,"accuracyPct":95.2},{"metric":"throughputRps","label":"Throughput","unit":"req/s","reference":488,"simulated":492.5,"deltaPct":0.9,"accuracyPct":99.1},{"metric":"errorRatePct","label":"Error Rate","unit":"%","reference":1.5,"simulated":1.5,"deltaPct":-0.1,"accuracyPct":100},{"metric":"costPerHour","label":"Cost / Hour","unit":"$/hr","reference":0.2302,"simulated":0.2314,"deltaPct":0.5,"accuracyPct":99.5}],"compositeScore":97.8,"comparisonStatus":"complete"},{"scenario":{"id":"burst","label":"Burst","description":"Traffic burst exceeding normal peak — flash sale, viral event, or coordinated load test.","trafficRps":1000,"referenceMetrics":{"latencyP50Ms":118,"latencyP95Ms":228,"cpuPct":68,"throughputRps":952,"errorRatePct":4.5,"costPerHour":0.2302},"sources":[{"metric":"cpuPct","citation":"DO s-4vcpu-8gb at 500 RPS/instance (13.9% of 3600-RPS capacity): real-world CPU ~64–72% on 4-vCPU Droplets. The AWS-calibrated formula over-predicts toward ~85% at this load due to capacity-basis mismatch, but less severely than with the smaller s-2vcpu-4gb. Source: DigitalOcean Droplet performance guidance.","url":"https://docs.digitalocean.com/products/droplets/resources/choose-plan/"},{"metric":"latencyP50Ms / latencyP95Ms","citation":"DO s-4vcpu-8gb under burst with DB approaching saturation: real-world P50 ~110–126 ms, P95 ~212–244 ms. Lower than s-2vcpu-4gb at equivalent load due to more vCPU headroom and a 200-connection DB vs the old 75-connection one. Source: DigitalOcean performance guidance.","url":"https://docs.digitalocean.com/products/droplets/resources/choose-plan/"},{"metric":"errorRatePct / throughputRps","citation":"DO db-s-2vcpu-4gb at 1000 RPS: ~200 concurrent connections = 100% pool utilisation, causing connection wait timeouts; real-world error rate ~4–5%. Effective throughput ~952 RPS (1000 × 0.952). Source: DigitalOcean Managed Databases documentation.","url":"https://docs.digitalocean.com/products/databases/mysql/"}]},"simulated":{"latencyP50Ms":118,"latencyP95Ms":230,"cpuPct":70.3,"throughputRps":954.999,"errorRatePct":4.5,"costPerHour":0.23137397260273967},"metrics":[{"metric":"latencyP50Ms","label":"P50 Latency","unit":"ms","reference":118,"simulated":118,"deltaPct":0,"accuracyPct":100},{"metric":"latencyP95Ms","label":"P95 Latency","unit":"ms","reference":228,"simulated":230,"deltaPct":0.9,"accuracyPct":99.1},{"metric":"cpuPct","label":"CPU Utilization","unit":"%","reference":68,"simulated":70.3,"deltaPct":3.4,"accuracyPct":96.6},{"metric":"throughputRps","label":"Throughput","unit":"req/s","reference":952,"simulated":955,"deltaPct":0.3,"accuracyPct":99.7},{"metric":"errorRatePct","label":"Error Rate","unit":"%","reference":4.5,"simulated":4.5,"deltaPct":0,"accuracyPct":100},{"metric":"costPerHour","label":"Cost / Hour","unit":"$/hr","reference":0.2302,"simulated":0.2314,"deltaPct":0.5,"accuracyPct":99.5}],"compositeScore":99,"comparisonStatus":"complete"}],"isCustomConfig":true}}}