Flower / Field notesReactive database · Raft · TypeScript

Goblin Pizza Express / Benchmark report

The rush-hour report.

Goblins bake. Drones deliver. Raft keeps the books.
Scheduled ovens, leased deliveries, and reactive shop totals under load.

Passed

2026-09-24 22:30:28 UTC

65.00 s total wall time · 0 unexpected failures

Customer goodput

11,129.2 / s

70% reads · 30% mutations among successful customer calls.

All-call tail latency

242.3 ms

All load-phase logical calls, including worker traffic, replays, discovery, and retries.

Orders delivered

96 / 96

244 pizzas · 1,708 copper in delivery revenue.

Quorum recovery

647.9 ms

First recorded crash to a serving quorum; restart catch-up is tracked separately.

Replica-local customer previews: replication lag is allowed. The selected pizza.shop.local method uses coherent locally applied snapshots without a quorum fence. Data, code, and aliases may lag without a bound; requests across replicas can return decreasing revisions. Replica-local is the benchmark default; choose --read-consistency fresh to measure quorum-fenced previews. Mutations and the final pizza.world audit retain fresh semantics.

The customer's view

Successful work, precisely counted.

One Raft group
Successful customer mix
70% reads / 30% mutations
Successful customer calls
669,240
Failed customer calls
0
Measured load duration
60.13 s
Load worker completions excluded
1,070
Load replay completions excluded
19,959
Customer goodput uses actual successful primary customer calls and the measured load duration.
Load measurementSuccessful completionsGoodput / sSuccessful customer share
Primary customer reads468,3027,787.770%
Primary customer mutations200,9383,341.530%
All primary customer calls669,24011,129.2100%
All logical calls, including workers and replays690,26911,478.9Includes non-customer traffic
Raw successful HTTP responses690,26911,478.9Includes retries and non-customer traffic

Counting definition: successful pizza.shop.local, pizza.order, and pizza.tip logical calls per measured load second. Transport retries count once when their customer call succeeds. Explicit .replay probes, worker claims and deliveries, polling, setup, audit, and every failed call contribute zero customer goodput. Actual successful read/mutation shares appear above.

The run verdict covers correctness and completion of any requested CPU profile. Customer goodput is an average over the recorded load phase; individual intervals can differ.

The rush, second by second

How the run unfolded

60 recorded time samples

Observed completion rate

Each point summarizes the interval since the previous sample.

Interval throughput over timeSuccessful HTTP responses and completed logical calls per second in each sampling interval. Dashed vertical lines mark recorded leader crashes. 60 samples over 60.26 s since load start. HTTP responses: first 14.9, last 8,300.8 events / s. Logical calls: first 14.9, last 8,300.8 events / s.00s4,65815s9,31530s13,97345s18,63060sLeader crash at 30.04 s after load startCrashHTTP responsesLogical calls
HTTP responsesLogical callsevents / s · elapsed load time

Cumulative HTTP tail latency

p95 of attempts since load began; not an interval percentile.

Cumulative HTTP p95 over timeThe latency p95 is cumulative from load start through each sample, including drain. It is not a moving-window or interval percentile. 60 samples over 60.26 s since load start. Cumulative p95: first 304.7, last 161.2 ms.00s7715s15430s23145s30860sLeader crash at 30.04 s after load startCrashCumulative p95
Cumulative p95ms · elapsed load time

Orders placed and delivered

Acknowledgements observed by the benchmark driver.

Acknowledged orders and deliveries over timeCumulative acknowledged order and delivery counts. Their gap shows business work still waiting for completion. 60 samples over 60.26 s since load start. Orders: first 0.0, last 96.0 orders. Deliveries: first 8.0, last 96.0 orders.00s2415s4830s7245s9660sLeader crash at 30.04 s after load startCrashOrdersDeliveries
OrdersDeliveriesorders · elapsed load time

Driver resident memory

Samples include the driver's runtime and retained workload records.

Driver resident memory over timeBenchmark driver resident memory sampled over elapsed load time. 60 samples over 60.26 s since load start. Driver RSS: first 367.7, last 202.0 MiB.00s9615s19230s28945s38560sLeader crash at 30.04 s after load startCrashDriver RSS
Driver RSSMiB · elapsed load time

Lines connect observed samples; no measurements are interpolated into the underlying data. Throughput is an interval rate, while p95 is cumulative. The driver and database share a machine. Any crash marker is positioned relative to the start of measured load.

Recorded time samples
Retries and HTTP failures are counts for each interval. Cumulative p95 covers all attempts since load began.
ElapsedPhaseHTTP attempts / sHTTP success / sLogical complete / sRetriesHTTP failuresCumulative p95OrdersDeliveriesDriver MiB
1.41 sload14.914.914.900304.7 ms08367.7
2.41 sload7,077.97,077.97,077.900307.7 ms9620371.3
3.41 sload13,058.013,058.013,058.000235.2 ms9640372.1
4.41 sload14,673.714,673.714,673.700179.8 ms9660372.8
5.41 sload12,341.812,341.812,341.800167.7 ms9678373.4
6.41 sload14,280.114,280.114,280.100158.0 ms9695374.2
7.41 sload11,783.411,783.411,783.400158.0 ms9696374.9
8.41 sload12,595.412,595.412,595.400156.4 ms9696375.5
9.44 sload18,630.218,630.218,630.200171.1 ms9696371.2
10.45 sload9,857.39,857.39,857.300176.3 ms9696372.2
11.45 sload9,900.89,900.89,900.800176.3 ms9696372.4
12.45 sload9,743.09,743.09,743.000179.8 ms9696373.3
13.45 sload10,454.810,454.810,454.800181.6 ms9696373.8
14.48 sload12,512.212,512.212,512.200176.3 ms9696375.2
15.48 sload13,877.213,877.213,877.200172.8 ms9696378.5
16.48 sload12,739.712,739.712,739.700171.1 ms9696379.2
17.48 sload13,360.413,360.413,360.400169.4 ms9696379.9
18.48 sload9,428.39,428.39,428.300171.1 ms9696380.7
19.48 sload10,091.410,091.410,091.400171.1 ms9696381.3
20.48 sload9,077.19,077.19,077.100174.5 ms9696382.0
21.49 sload10,520.610,520.610,520.600174.5 ms9696382.7
22.49 sload9,843.49,843.49,843.400176.3 ms9696383.3
23.49 sload10,067.410,067.410,067.400176.3 ms9696384.0
24.49 sload9,145.69,145.69,145.600176.3 ms9696384.7
25.49 sload10,145.110,145.110,145.100176.3 ms9696363.2
26.50 sload10,815.910,815.910,815.900176.3 ms9696363.1
27.50 sload11,240.111,240.111,240.100178.0 ms9696358.5
28.50 sload15,082.415,082.415,082.400176.3 ms9696333.2
29.50 sload14,157.614,157.614,157.600174.5 ms9696305.9
30.50 sload12,461.19,807.89,807.82,1402,652172.8 ms9696309.8
31.50 sload10,089.08,804.58,804.51,7561,285172.8 ms9696311.7
32.50 sload13,336.812,943.812,943.8399393172.8 ms9696313.9
33.50 sload13,192.512,918.012,918.0310275171.1 ms9696315.8
34.50 sload10,996.510,758.810,758.8238238171.1 ms9696317.3
35.51 sload1,860.61,857.61,857.633171.1 ms9696319.0
36.51 sload8,458.38,458.38,458.300171.1 ms9696320.4
37.51 sload11,754.011,754.011,754.000171.1 ms9696321.9
38.51 sload6,857.36,857.36,857.300172.8 ms9696322.4
39.51 sload14,398.514,398.514,398.500171.1 ms9696324.0
40.51 sload12,035.012,035.012,035.000169.4 ms9696325.4
41.51 sload13,604.313,604.313,604.300169.4 ms9696326.8
42.51 sload12,178.412,178.412,178.400167.7 ms9696327.8
43.51 sload16,129.816,129.816,129.800167.7 ms9696328.6
44.51 sload15,450.815,450.815,450.800167.7 ms9696326.5
45.51 sload13,979.913,979.913,979.900166.0 ms9696326.0
46.52 sload5,038.75,038.75,038.700166.0 ms9696322.6
47.52 sload12,257.012,257.012,257.000166.0 ms9696323.5
48.52 sload12,914.612,914.612,914.600164.4 ms9696324.2
49.52 sload13,469.113,469.113,469.100164.4 ms9696325.7
50.52 sload11,526.211,526.211,526.200164.4 ms9696327.2
51.52 sload14,619.314,619.314,619.300162.8 ms9696328.2
52.52 sload15,554.915,554.915,554.900162.8 ms9696329.5
53.52 sload13,928.913,928.913,928.900161.2 ms9696331.2
54.52 sload14,461.214,461.214,461.200161.2 ms9696332.4
55.52 sload7,449.67,449.67,449.600161.2 ms9696316.8
56.52 sload12,132.912,132.912,132.900161.2 ms9696193.4
57.52 sload12,417.312,417.312,417.300161.2 ms9696192.9
58.52 sload14,408.914,408.914,408.900161.2 ms9696197.0
59.52 sload14,034.414,034.414,034.400159.6 ms9696198.1
60.26 sdrain8,300.88,300.88,300.800161.2 ms9696202.0

The shape of the rush

Throughput & response time

Closed-loop workload

Successful work by phase

Actual measured phase durations appear below the bars.

Successful throughput by phaseSuccessful HTTP responses and completed logical calls per second for each phase. Setup: HTTP responses 1.7 events / s, Logical calls 1.7 events / s. Warmup: HTTP responses 15,219.8 events / s, Logical calls 15,219.8 events / s. Load: HTTP responses 11,478.9 events / s, Logical calls 11,478.9 events / s. Drain: HTTP responses 8.1 events / s, Logical calls 8.1 events / s. Audit: HTTP responses 33.9 events / s, Logical calls 33.9 events / s03,8057,61011,41515,220Setup · HTTP responses: 1.7 events / sSetup · Logical calls: 1.7 events / sSetup2.90 sWarmup · HTTP responses: 15,219.8 events / sWarmup · Logical calls: 15,219.8 events / sWarmup1.01 sLoad · HTTP responses: 11,478.9 events / sLoad · Logical calls: 11,478.9 events / sLoad60.13 sDrain · HTTP responses: 8.1 events / sDrain · Logical calls: 8.1 events / sDrain0.12 sAudit · HTTP responses: 33.9 events / sAudit · Logical calls: 33.9 events / sAudit0.03 s
HTTP responsesLogical callsevents / s · linear scale

Logical latency by phase

Retry and discovery time remain visible in the tails.

Logical latency percentiles by phaseLogical call p50, p95, and p99. Quantile bars are summary measurements, not a latency distribution. Setup: p50 235.2 ms, p95 260.2 ms, p99 260.2 ms. Warmup: p50 29.1 ms, p95 67.8 ms, p99 86.1 ms. Load: p50 7.8 ms, p95 161.2 ms, p99 242.3 ms. Drain: p50 7.3 ms, p95 7.3 ms, p99 7.3 ms. Audit: p50 2.3 ms, p95 11.3 ms, p99 11.3 ms065130195260Setup · p50: 235.2 msSetup · p95: 260.2 msSetup · p99: 260.2 msSetupWarmup · p50: 29.1 msWarmup · p95: 67.8 msWarmup · p99: 86.1 msWarmupLoad · p50: 7.8 msLoad · p95: 161.2 msLoad · p99: 242.3 msLoadDrain · p50: 7.3 msDrain · p95: 7.3 msDrain · p99: 7.3 msDrainAudit · p50: 2.3 msAudit · p95: 11.3 msAudit · p99: 11.3 msAudit
p50p95p99ms · linear scale

A successful HTTP response is not necessarily a new business mutation: receipt replays count as responses, and queries also count. Phase percentiles cover successful and failed attempts or calls. Warmup uses reads and is not comparable to the mixed load phase.

Every call accounted for

Requests, retries & methods

Load details expanded below
Load HTTP attempts
695,115
Load HTTP successes
690,269
Load HTTP failures
4,846
Load retry attempts
4,846
Load duplicate responses
19,959
Load logical failures
0

Network retries belong to the same logical call. Deliberate deduplication probes appear separately as .replay calls. Expected stale-lease failures can appear in these measurements without failing the benchmark.

Setup2.90 s · 5 HTTP attempts
Successful HTTP / s
1.7
Completed logical / s
1.7
HTTP failures
0
Logical failures
0

HTTP error categories: None recorded

Setup HTTP attempts. All latency columns are milliseconds.
HTTP methodAttemptsSucceededFailedRetriesDuplicatesSuccess / sp50p95p99Max
deploy110000.3260.0 ms260.0 ms260.0 ms260.0 ms
pizza.setup110000.326.0 ms26.0 ms26.0 ms26.0 ms
pizza.shop330001.0234.7 ms234.7 ms234.7 ms234.7 ms
Logical calls, including explicit replays
One logical call includes discovery, retries, and backoff. Explicit replay checks are separate .replay calls.
Logical methodCallsCompletedFailedDuplicatesComplete / sp50p95p99Max
deploy11000.3260.2 ms260.2 ms260.2 ms260.2 ms
pizza.setup11000.326.1 ms26.1 ms26.1 ms26.1 ms
pizza.shop33001.0234.7 ms234.7 ms234.7 ms234.7 ms
Warmup1.01 s · 15,322 HTTP attempts
Successful HTTP / s
15,219.8
Completed logical / s
15,219.8
HTTP failures
0
Logical failures
0

HTTP error categories: None recorded

Warmup HTTP attempts. All latency columns are milliseconds.
HTTP methodAttemptsSucceededFailedRetriesDuplicatesSuccess / sp50p95p99Max
pizza.shop.local15,32215,32200015,219.829.1 ms67.8 ms86.1 ms114.9 ms
Logical calls, including explicit replays
One logical call includes discovery, retries, and backoff. Explicit replay checks are separate .replay calls.
Logical methodCallsCompletedFailedDuplicatesComplete / sp50p95p99Max
pizza.shop.local15,32215,3220015,219.829.1 ms67.8 ms86.1 ms114.9 ms
Load60.13 s · 695,115 HTTP attempts
Successful HTTP / s
11,478.9
Completed logical / s
11,478.9
HTTP failures
4,846
Logical failures
0

HTTP error categories: HTTP_503 3 · network 4,843

Load HTTP attempts. All latency columns are milliseconds.
HTTP methodAttemptsSucceededFailedRetriesDuplicatesSuccess / sp50p95p99Max
pizza.claim97497400016.279.5 ms185.2 ms379.2 ms754.4 ms
pizza.deliver11111100151.881.9 ms196.6 ms265.0 ms302.8 ms
pizza.order1041040081.7200.6 ms408.0 ms408.0 ms408.0 ms
pizza.shop.local469,564468,3021,2621,26207,787.73.2 ms32.8 ms65.2 ms1,740.7 ms
pizza.tip224,362220,7783,5843,58419,9363,671.5101.0 ms208.7 ms307.7 ms808.6 ms
Logical calls, including explicit replays
One logical call includes discovery, retries, and backoff. Explicit replay checks are separate .replay calls.
Logical methodCallsCompletedFailedDuplicatesComplete / sp50p95p99Max
pizza.claim9749740016.279.5 ms189.0 ms611.4 ms754.5 ms
pizza.deliver9696001.681.9 ms210.8 ms302.8 ms302.8 ms
pizza.deliver.replay15150150.278.7 ms167.1 ms167.1 ms167.1 ms
pizza.order9696001.6200.6 ms259.3 ms259.3 ms259.3 ms
pizza.order.replay88080.1408.0 ms408.0 ms408.0 ms408.0 ms
pizza.shop.local468,302468,302007,787.73.2 ms32.8 ms66.5 ms1,740.7 ms
pizza.tip200,842200,842003,339.9102.0 ms212.9 ms346.7 ms844.3 ms
pizza.tip.replay19,93619,936019,936331.5100.0 ms187.1 ms410.6 ms840.3 ms
Drain0.12 s · 1 HTTP attempts
Successful HTTP / s
8.1
Completed logical / s
8.1
HTTP failures
0
Logical failures
0

HTTP error categories: None recorded

Drain HTTP attempts. All latency columns are milliseconds.
HTTP methodAttemptsSucceededFailedRetriesDuplicatesSuccess / sp50p95p99Max
pizza.world110008.17.2 ms7.2 ms7.2 ms7.2 ms
Logical calls, including explicit replays
One logical call includes discovery, retries, and backoff. Explicit replay checks are separate .replay calls.
Logical methodCallsCompletedFailedDuplicatesComplete / sp50p95p99Max
pizza.world11008.17.3 ms7.3 ms7.3 ms7.3 ms
Audit0.03 s · 9 HTTP attempts
Successful HTTP / s
33.9
Completed logical / s
33.9
HTTP failures
8
Logical failures
8

HTTP error categories: HTTP_422 8

Audit HTTP attempts. All latency columns are milliseconds.
HTTP methodAttemptsSucceededFailedRetriesDuplicatesSuccess / sp50p95p99Max
pizza.deliver808000.02.2 ms2.7 ms2.7 ms2.7 ms
pizza.world1100033.911.1 ms11.1 ms11.1 ms11.1 ms
Logical calls, including explicit replays
One logical call includes discovery, retries, and backoff. Explicit replay checks are separate .replay calls.
Logical methodCallsCompletedFailedDuplicatesComplete / sp50p95p99Max
pizza.deliver80800.02.3 ms2.8 ms2.8 ms2.8 ms
pizza.world110033.911.3 ms11.3 ms11.3 ms11.3 ms

From the oven to the doorstep

Business completion latency

Load and drain combined
Oven lateness and order-to-delivery percentilesOven lateness is callback execution time minus its due time. Order-to-delivery includes baking, queueing, abandoned leases, and delivery. Oven lateness: p50 444.7 ms, p95 1,238.0 ms, p99 1,238.0 ms. Order to door: p50 3,157.5 ms, p95 5,350.3 ms, p99 5,977.0 ms01,4942,9894,4835,977Oven lateness · p50: 444.7 msOven lateness · p95: 1,238.0 msOven lateness · p99: 1,238.0 msOven lateness96 samplesOrder to door · p50: 3,157.5 msOrder to door · p95: 5,350.3 msOrder to door · p99: 5,977.0 msOrder to door96 samples
p50p95p99ms · linear scale
Configured oven delay
250.0 ms
Configured lease duration
2,000.0 ms
Delivered orders / s, including drain
1.6
Retained order cap
96

Oven lateness measures how far execution runs past the requested deadline. It does not include the configured baking delay. Order-to-door latency includes the entire lifecycle. These are recorded quantiles; raw histogram buckets are not present in this report.

Approximate nearest-rank quantiles. Min and max are exact observed values.
MeasurementSamplesMinp50p95p99MaxInvalid / overflow
Oven lateness9632.0 ms444.7 ms1,238.0 ms1,238.0 ms1,238.0 ms0 / 0
Order to delivery96537.0 ms3,157.5 ms5,350.3 ms5,977.0 ms5,977.0 ms0 / 0

The goblin ledger

Business invariants & tenant rankings

Audit passed
96 of 96 acknowledged orders delivered. The independent audit found no invariant violations.
Scores are revenue plus tips, in integer copper coins. Rows follow each tenant’s independently derived leaderboard; this is not a global ranking.
Tenant / kitchenOrdersPizzas deliveredStock leftRevenueTipsScore
tenant-2 / The Crispy Cauldron · store-039105289735170,288171,023
tenant-2 / Dough or Die · store-3243902810,31910,347
tenant-2 / The Sizzling Slime · store-24133819110,19410,285
tenant-2 / Mushroom Mayhem · store-1253893510,02710,062
tenant-3 / The Goblin's Slice · store-043100294700170,652171,352
tenant-3 / Dragon Breath Delivery · store-14113837710,39810,475
tenant-3 / The Crispy Cauldron · store-2263884210,13010,172
tenant-3 / Mushroom Mayhem · store-30039409,8589,858

A dragon ate the leader

Failover, leases & replay safety

3 Raft replicas · one group

Leader crash 1

2026-09-24 22:31:02 UTC
Crash recovery: leader 1 killed; serving quorum after 647.9 ms; restarted node caught up after 4,808.2 ms. Both durations are measured from the crash.Leader 1 killed0.0 msQuorum serving; leader 3647.9 msRestarted node caught up4,808.2 ms
Recovery observations, all measured from the leader crash
Recovery milestoneTime since crash
Candidate observed in metrics535.6 ms
Leader observed in metrics564.0 ms
Quorum read probe completed647.8 ms
Leader discovery returned647.9 ms
Restarted node caught up4,808.2 ms

Metrics observations follow polling and are not exact internal transition times. A brief candidate state can be missed; observing a leader alone does not establish a serving quorum. Missing milestones were not recorded or did not finish. Recovery and catch-up durations share the crash as time zero.

Receipt replays verified
19,959
Stale drones rejected by audit
8
Leases claimed
104
Empty claims
870
Deliberately abandoned leases
8
Reclaimed jobs
8
Lease losses during workload
0
Orders issued
96
End-of-run replica observations, collected separately. These are replicas of one Raft group, not independent shards.
ReplicaObserved roleObserved leaderLast applied index
Node 1Follower31,450
Node 2Follower31,450
Node 3Leader31,450

The cost of a pizza rush

Resource footprint

Co-located processes

Sampled peak resident memory

Sampled peak resident memory by processPeak observed resident memory per server and benchmark driver. Samples can miss brief peaks; per-process peaks need not occur together. Node 1: Peak sampled RSS 435.3 MiB. Node 2: Peak sampled RSS 401.3 MiB. Node 3: Peak sampled RSS 420.2 MiB. Driver: Peak sampled RSS 384.7 MiB0109218326435Node 1 · Peak sampled RSS: 435.3 MiBNode 1Node 2 · Peak sampled RSS: 401.3 MiBNode 2Node 3 · Peak sampled RSS: 420.2 MiBNode 3Driver · Peak sampled RSS: 384.7 MiBDriver
Peak sampled RSSMiB · linear scale
Cluster disk footprint
531.06 MiB
Node controller sampled load cores
0.06
Node controller CPU, user / entire run
4,429.4 ms
Node controller CPU, system / entire run
584.7 ms
Node controller event-loop delay p99
45.9 ms
Node controller sampled peak RSS
384.7 MiB
Rust customer driver peak RSS
24.8 MiB
Rust customer driver sampled load cores
0.23

RSS is sampled, not an allocation total. Server CPU, when available, uses cumulative process CPU deltas over sampled load intervals; one core means one CPU-second per elapsed second. Restart and phase-boundary gaps are excluded. Older reports did not measure it. The load generator and all nodes share the same host, so these numbers include local resource contention and do not establish multi-host network performance.

CPU observations during load; node coverage can differ. Includes all process threads.
ServerMean CPU coresCPU time coveredWall time coveredIntervals
10.5431,330.0 ms57,523.6 ms57
20.2011,610.0 ms59,523.7 ms59
30.5230,690.0 ms59,523.7 ms59

Make the result reproducible

Workload, host & measurement limits

Report schema 1

Workload configuration

Application transport
HTTP/2 · h2c prior knowledge
Admin and discovery transport
HTTP/1.1
Query routing
Round-robin across live replicas
Customer read policy
Replica-local · lag allowed
Bundle initialization
static
Customer loops / drone loops
512 / 4
Requested load / warmup
60.0 s / 1.0 s
Maximum drain
60.0 s
Tenants / stores per tenant
2 / 4
Hot stores per tenant
1
Hot-set mixture weight
80%
Duplicate / abandonment probability
10% / 10%
Worker polling interval
150.0 ms
Request timeout / retry budget
2,000.0 ms / 15,000.0 ms
Workload seed
42:group-1
Crash injection requested
Yes

Execution environment

JavaScript engine
quickjs
Runtime settings
{"TOKIO_WORKER_THREADS":"2","FLOWER_WRITER_BATCH_MODE":"adaptive","FLOWER_WRITER_QUEUE_CAPACITY":"1024","FLOWER_QUERY_WORKERS":"16","FLOWER_PREPARATION_WORKERS":"16","FLOWER_WRITER_PREPARATION_WORKERS":"1","FLOWER_WRITER_BATCH_MS":"200","FLOWER_OTEL_ENABLED":"0","OTEL_SDK_DISABLED":"true"}
CPU
Apple M5 Pro
Logical CPUs
18
Operating system
darwin 27.2.0
Architecture
arm64
Node.js controller
v26.10.0
Customer driver
rust
Customer driver SHA-256
1cf843cbf4e064d13a1d3597d9295c509325fee5c74b6eb2242ac7e98ece553d
Co-located database nodes
3
Binary SHA-256
3b47cf25cfa2a278f6b96e9ab675c70694b1ad4408e6f09a18e2555c0939c9c9
Binary size
44,575,760 bytes
Started
2026-09-24 22:30:28 UTC
Finished
2026-09-24 22:31:33 UTC
Preserved database files
Not retained

How to read these numbers

Complete run options and bundle identity
{
  "options": {
    "offeredRate": 0,
    "groups": 1,
    "tenants": 2,
    "duration": 60,
    "warmup": 1,
    "drain": 60,
    "concurrency": 512,
    "workers": 4,
    "shops": 4,
    "hotShops": 1,
    "hotProbability": 0.8,
    "maxOrders": 96,
    "bakeMs": 250,
    "leaseMs": 2000,
    "duplicateRate": 0.1,
    "abandonRate": 0.1,
    "pollMs": 150,
    "requestTimeoutMs": 2000,
    "retryBudgetMs": 15000,
    "nodes": 3,
    "seed": "42:group-1",
    "initialization": "static",
    "queryRouting": "replicas",
    "readConsistency": "replica-local",
    "driver": "rust",
    "driverBinary": "/Users/pcarrier/src/flower/target/release/flower-bench-driver",
    "binary": "/Users/pcarrier/src/flower/target/release/flower",
    "json": "/Users/pcarrier/src/flower/bench/results/latest-groups/group-1.json",
    "html": "/Users/pcarrier/src/flower/bench/results/latest-groups/group-1.html",
    "baseline": null,
    "cpuProfile": null,
    "chaos": true,
    "http2": true,
    "keepData": false,
    "help": false,
    "runId": "5a967ca9-9901-46a1-a34c-9d40e82dfc3f",
    "tenantIds": [
      "tenant-2",
      "tenant-3"
    ]
  },
  "runtime": {
    "engine": "quickjs",
    "settings": {
      "TOKIO_WORKER_THREADS": "2",
      "FLOWER_WRITER_BATCH_MODE": "adaptive",
      "FLOWER_WRITER_QUEUE_CAPACITY": "1024",
      "FLOWER_QUERY_WORKERS": "16",
      "FLOWER_PREPARATION_WORKERS": "16",
      "FLOWER_WRITER_PREPARATION_WORKERS": "1",
      "FLOWER_WRITER_BATCH_MS": "200",
      "FLOWER_OTEL_ENABLED": "0",
      "OTEL_SDK_DISABLED": "true"
    }
  },
  "queryRouting": {
    "mode": "replicas",
    "consistency": "replica-local",
    "auditConsistency": "fresh",
    "nodes": [
      {
        "id": 1,
        "url": "http://127.0.0.1:51187",
        "attempts": 152151,
        "completed": 150889,
        "failures": 1262
      },
      {
        "id": 2,
        "url": "http://127.0.0.1:51202",
        "attempts": 166370,
        "completed": 166370,
        "failures": 0
      },
      {
        "id": 3,
        "url": "http://127.0.0.1:51203",
        "attempts": 166370,
        "completed": 166370,
        "failures": 0
      }
    ]
  },
  "binary": {
    "path": "/Users/pcarrier/src/flower/target/release/flower",
    "sha256": "3b47cf25cfa2a278f6b96e9ab675c70694b1ad4408e6f09a18e2555c0939c9c9",
    "bytes": 44575760,
    "modifiedAt": "2026-09-24T22:08:02.762Z"
  },
  "driver": {
    "kind": "rust",
    "orchestration": "node",
    "binary": {
      "path": "/Users/pcarrier/src/flower/target/release/flower-bench-driver",
      "sha256": "1cf843cbf4e064d13a1d3597d9295c509325fee5c74b6eb2242ac7e98ece553d",
      "bytes": 7136160,
      "modifiedAt": "2026-09-24T22:07:35.617Z"
    },
    "warmupRequests": 3
  },
  "bundleHash": "264f2718f2336ea92b850ec939282695609e4ac929a73006455bf212872473b0"
}

When the oven smokes

Diagnostics

0 unexpected failures

No unexpected failure details were recorded. Intentional stale-lease rejections and retried HTTP failures are reported in their own counters.