Benchmark Report
High disk latency · Disk Performance
The worst 4K direct-IO operations take longer than a tick's whole budget. Chunk saves stall the main thread.
- direct_read_4k_p99_us = 5,178.62 us
- direct_write_4k_p99_us = 1,571.53 us
- latency_budget_us = 1,000 us
Weakest test: Mob AI, 5.09 TPS on average. The 15 other tests hold at least 10.62 TPS.
Compared with other servers
Every measured server counts here, including private, anonymous and expired reports. Each server counts once, with its best run.
This report scores better than
0%
of the 6 servers running Paper & forks on Minecraft 26.3
Total score 31,356 · group median 103,836
Bottom quarter- Single Core
Better than 0% · 49,120 vs median 151,189 · Bottom quarter
- Gameplay
Better than 0% · 46,801 vs median 66,126 · Bottom quarter
- Exploration
Better than 0% · 51,092 vs median 80,634 · Bottom quarter
- Entities
Better than 0% · 47,106 vs median 75,282 · Bottom quarter
- Mechanics
Better than 0% · 43,507 vs median 46,380 · Bottom quarter
- Hardware
Better than 0% · 18,166 vs median 48,218 · Bottom quarter
- Memory
Better than 0% · 9,666 vs median 30,056 · Bottom quarter
- Disk
Better than 0% · 3,489 vs median 55,398 · Bottom quarter
- Multi-Core
Better than 12% · 49,767 vs median 91,322 · Bottom quarter
Test by test
Each tick test's average MSPT (lower is better) and TPS, against the same group, each server with the same run as above. A test shows when at least 5 servers of the group passed it. The bar follows the MSPT.
- Block Physics
Better than 0% · MSPT 29 ms vs median 10 ms · TPS 20 vs median 20 · Bottom quarter
- Bone Meal Growth
Better than 34% · MSPT 1.5 ms vs median 1.3 ms · TPS 20 vs median 20
- Chunk Loading
Better than 0% · MSPT 12.5 ms vs median 4.2 ms · TPS 18.4 vs median 19.7 · Bottom quarter
- Chunk Ticking
Better than 10% · MSPT 39.7 ms vs median 15 ms · TPS 19.6 vs median 20 · Bottom quarter
- Explosion
Better than 10% · MSPT 6.7 ms vs median 3.3 ms · TPS 19.8 vs median 19.9 · Bottom quarter
- Hopper
Better than 8% · MSPT 4.4 ms vs median 1.9 ms · TPS 19.6 vs median 20 · Bottom quarter
- Lighting Update
Better than 15% · MSPT 3.3 ms vs median 1.7 ms · TPS 20 vs median 20 · Bottom quarter
- Mob AI
Better than 0% · MSPT 445.8 ms vs median 146.1 ms · TPS 5.1 vs median 8.2 · Bottom quarter
- Redstone
Better than 0% · MSPT 97 ms vs median 22.4 ms · TPS 19.3 vs median 17.6 · Bottom quarter
- Ticking Tile Entities
Better than 4% · MSPT 7.3 ms vs median 2.4 ms · TPS 20 vs median 20 · Bottom quarter
- World Save
Better than 68% · MSPT 0.8 ms vs median 1.1 ms · TPS 20 vs median 20
Scored with version 8.3.0 of the scoring rules. Groups of fewer than 5 servers are never used. Updated Oct 28, 2026, 12:40 PM UTC. Community trends · All measurements by CPU
What matters
Read from the measurements, never estimatedHigh disk latency
The worst 4K direct-IO operations take longer than a tick's whole budget. Chunk saves stall the main thread.
Likely cause: disk latency
- direct_read_4k_p99_us = 5,178.62 us
- direct_write_4k_p99_us = 1,571.53 us
- latency_budget_us = 1,000 us
Slow random reads
Small random reads are far below what a server NVMe holds. Chunk loading and world saves pay for it first.
Likely cause: random reads
- direct_read_4k_iops = 3,668.96 IOPS
- reference_iops = 40,000 IOPS
Non-default server configuration
Compare with care against a server running default settings.
- view-distance 6 (default 10)
- simulation-distance 4 (default 10)
›+12 more findings
chunkLoadingThe worst ticks are several times the mean. Average TPS stays fine while players feel the stutter.
- mspt_avg = 12.5 ms
- mspt_p99 = 66.26 ms
- spike_ratio = 5.3 ratio
- sample_count = 111 count
- spike_ticks = 3 count
- spike_window_ticks = 123 count
- spike_threshold_ms = 287.5 ms
- spike_ticks_with_gc = 0 count
- spike_ticks_with_throttle = 1 count
- spike_ticks_with_steal = 2 count
- spike_tick_ms = 1,318.8 ms
- spike_gc_pause_ms = 0 ms
- spike_throttled_ms = 0.5 ms
- spike_sample_interval_ms = 40 ms
- explained_share_threshold = 0.5 ratio
- explained_time_share_threshold = 0.25 ratio
mobAIThe worst ticks are several times the mean. Average TPS stays fine while players feel the stutter.
- mspt_avg = 445.8 ms
- mspt_p99 = 1,764.61 ms
- spike_ratio = 3.96 ratio
- sample_count = 55 count
- spike_ticks = 15 count
- spike_window_ticks = 57 count
- spike_threshold_ms = 573.1 ms
- spike_ticks_with_gc = 0 count
- spike_ticks_with_throttle = 1 count
- spike_ticks_with_steal = 5 count
- spike_tick_ms = 13,801.9 ms
- spike_gc_pause_ms = 0 ms
- spike_throttled_ms = 1 ms
- spike_sample_interval_ms = 40 ms
- explained_share_threshold = 0.5 ratio
- explained_time_share_threshold = 0.25 ratio
networkThe worst ticks are several times the mean. Average TPS stays fine while players feel the stutter.
- mspt_avg = 127.7 ms
- mspt_p99 = 570.27 ms
- spike_ratio = 4.47 ratio
- sample_count = 161 count
- spike_ticks = 44 count
- spike_window_ticks = 420 count
- spike_threshold_ms = 155.9 ms
- spike_ticks_with_gc = 0 count
- spike_ticks_with_throttle = 0 count
- spike_ticks_with_steal = 7 count
- spike_tick_ms = 10,820.3 ms
- spike_gc_pause_ms = 0 ms
- spike_throttled_ms = 0 ms
- spike_sample_interval_ms = 40 ms
- explained_share_threshold = 0.5 ratio
- explained_time_share_threshold = 0.25 ratio
chunkLoadingThe average tick rate is healthy but the worst moments fall to a standstill. It is regularity that is missing, not power.
- tps_avg = 18.42 tps
- tps_p1 = 10 tps
- sample_count = 111 count
redstoneThe average tick rate is healthy but the worst moments fall to a standstill. It is regularity that is missing, not power.
- tps_avg = 19.34 tps
- tps_p1 = 5 tps
- sample_count = 61 count
redstonesingle coreOne game thread is already over its tick budget under the benchmark's own load. Players and redstone will push it further.
- mspt_avg = 97 ms
- mspt_target = 35 ms
- tps_avg = 19.34 tps
- tps_held_threshold = 18 tps
chunkLoadingThe reference computation, timed just before this test and just before the next one, ran markedly slower than the run's reference, the best of its start, middle and end timings (or the hypervisor stole CPU time during the test, or, for the CPU test, the machine had more tasks ready to run than processors): this test did not get the processor the rest of the run got. For a scored benchmark, the server's run statistics leave this run out of the scores this test feeds, and keep it for the others when their own tests ran in stable windows. The run's own verdict does not change.
- window_slowdown_percent = 19.25 %
- threshold_percent = 15 %
- probe_basis_median = 1
- probe_noise_percent = 7.91 %
- reference_best_ms = 28.83 ms
- opening_best_ms = 27.33 ms
- opening_judged_ms = 28.55 ms
- closing_best_ms = 28.08 ms
- closing_judged_ms = 34.38 ms
- steal_percent = 0.09 %
- steal_threshold_percent = 5 %
hopperThe reference computation, timed just before this test and just before the next one, ran markedly slower than the run's reference, the best of its start, middle and end timings (or the hypervisor stole CPU time during the test, or, for the CPU test, the machine had more tasks ready to run than processors): this test did not get the processor the rest of the run got. For a scored benchmark, the server's run statistics leave this run out of the scores this test feeds, and keep it for the others when their own tests ran in stable windows. The run's own verdict does not change.
- window_slowdown_percent = 19.25 %
- threshold_percent = 15 %
- probe_basis_median = 1
- probe_noise_percent = 7.91 %
- reference_best_ms = 28.83 ms
- opening_best_ms = 28.08 ms
- opening_judged_ms = 34.38 ms
- closing_best_ms = 27.02 ms
- closing_judged_ms = 29.07 ms
- steal_percent = 0.07 %
- steal_threshold_percent = 5 %
lightingUpdateThe reference computation, timed just before this test and just before the next one, ran markedly slower than the run's reference, the best of its start, middle and end timings (or the hypervisor stole CPU time during the test, or, for the CPU test, the machine had more tasks ready to run than processors): this test did not get the processor the rest of the run got. For a scored benchmark, the server's run statistics leave this run out of the scores this test feeds, and keep it for the others when their own tests ran in stable windows. The run's own verdict does not change.
- window_slowdown_percent = 20.71 %
- threshold_percent = 15 %
- probe_basis_median = 1
- probe_noise_percent = 7.91 %
- reference_best_ms = 28.83 ms
- opening_best_ms = 27.48 ms
- opening_judged_ms = 28.92 ms
- closing_best_ms = 28.51 ms
- closing_judged_ms = 34.8 ms
- steal_percent = 0.09 %
- steal_threshold_percent = 5 %
memoryThe reference computation, timed just before this test and just before the next one, ran markedly slower than the run's reference, the best of its start, middle and end timings (or the hypervisor stole CPU time during the test, or, for the CPU test, the machine had more tasks ready to run than processors): this test did not get the processor the rest of the run got. For a scored benchmark, the server's run statistics leave this run out of the scores this test feeds, and keep it for the others when their own tests ran in stable windows. The run's own verdict does not change.
- window_slowdown_percent = 16.75 %
- threshold_percent = 15 %
- probe_basis_median = 1
- probe_noise_percent = 7.91 %
- reference_best_ms = 28.83 ms
- opening_best_ms = 28.09 ms
- opening_judged_ms = 33.66 ms
- closing_best_ms = 27.33 ms
- closing_judged_ms = 28.55 ms
- steal_percent = 0.06 %
- steal_threshold_percent = 5 %
networkThe reference computation, timed just before this test and just before the next one, ran markedly slower than the run's reference, the best of its start, middle and end timings (or the hypervisor stole CPU time during the test, or, for the CPU test, the machine had more tasks ready to run than processors): this test did not get the processor the rest of the run got. For a scored benchmark, the server's run statistics leave this run out of the scores this test feeds, and keep it for the others when their own tests ran in stable windows. The run's own verdict does not change.
- window_slowdown_percent = 16.75 %
- threshold_percent = 15 %
- probe_basis_median = 1
- probe_noise_percent = 7.91 %
- reference_best_ms = 28.83 ms
- opening_best_ms = 28.38 ms
- opening_judged_ms = 30.81 ms
- closing_best_ms = 28.09 ms
- closing_judged_ms = 33.66 ms
- steal_percent = 0.09 %
- steal_threshold_percent = 5 %
tickingTileEntityThe reference computation, timed just before this test and just before the next one, ran markedly slower than the run's reference, the best of its start, middle and end timings (or the hypervisor stole CPU time during the test, or, for the CPU test, the machine had more tasks ready to run than processors): this test did not get the processor the rest of the run got. For a scored benchmark, the server's run statistics leave this run out of the scores this test feeds, and keep it for the others when their own tests ran in stable windows. The run's own verdict does not change.
- window_slowdown_percent = 20.71 %
- threshold_percent = 15 %
- probe_basis_median = 1
- probe_noise_percent = 7.91 %
- reference_best_ms = 28.83 ms
- opening_best_ms = 28.51 ms
- opening_judged_ms = 34.8 ms
- closing_best_ms = 29.03 ms
- closing_judged_ms = 31.07 ms
- steal_percent = 0.08 %
- steal_threshold_percent = 5 %
A finding stays silent rather than attribute a cause without evidence. An absent cause is an answer, not an omission.
Test by test
Problems first — select a row for the full test- Mob AIEntities5.09UnstableCritical
- Network (NBT)Network10.62UnstableCritical
- Chunk LoadingExploration18.42UnstableCritical
- Disk PerformanceMemory & Storage20.00StableCritical
- RedstoneSingle-Core Stress19.34effective 12.1UnstableUnstable
- Chunk TickingExploration19.61ModerateUnstable
- HopperMechanics19.61StableUnstable
- ExplosionMechanics19.81StableUnstable
- Block PhysicsSingle-Core Stress20.00UnstableHealthy
- Bone Meal GrowthMechanics20.00StableHealthy
- Lighting UpdateExploration20.00StableHealthy
- Memory PerformanceMemory & Storage20.00StableHealthy
- Multi-Core CPUCPU20.00StableHealthy
- Single-Core (tick-aligned)CPU20.00StableHealthy
- Ticking Tile EntitiesMechanics20.00StableHealthy
- World SaveMechanics20.00StableHealthy
Settings your measurements justify
Each setting below is triggered by a measurement from this report, and names the test to replay to find out whether it helped.
Shrink the monster activation range
world-settings.default.entity-activation-range.monsters: 24Measured : mobAI.mspt.avg 445.8 ms (threshold 35.0 ms)
Mob AI costs more per tick than the budget allows. The activation range decides how far from a player a monster keeps thinking; shrinking it puts distant ones to sleep without removing them.
Test to replay : mobAI
Lower the monster spawn limit
spawn-limits.monsters: 50Currently : 70
Measured : mobAI.mspt.avg 445.8 ms (threshold 50.0 ms)
Mob AI is past the critical cost per tick, where shrinking the activation range is no longer enough. The spawn limit caps how many monsters exist per player at all, which is the only setting here that reduces the work rather than deferring it.
Test to replay : mobAI
These settings trade a little gameplay fidelity for tick budget. Apply one at a time and replay the named test: the effect stays a hypothesis until you have measured it.
Briefing for a language model
Paste this into the assistant of your choice to have it rewritten in prose. It already carries every number, so the model has none left to invent.
{
"format": "voxelbench.tuning.v1",
"instruction": "Rewrite each recommendation below as plain prose for a server operator. Every number you are allowed to use is already in this document: the measured value, its threshold, and the proposed setting. Do not compute, estimate or predict any other number, and never state an expected gain. Keep every setting name and value exactly as written.",
"server": {
"type": "paper",
"minecraft": "26.3",
"java": "25.0.5.0.3"
},
"recommendations": [
{
"id": "spigot.portee-activation-monstres",
"family": "spigot",
"measure": "mobAI.mspt.avg",
"value": 445.8,
"unit": "ms",
"threshold": 35,
"control_test": "mobAI",
"file": "spigot.yml",
"setting": "world-settings.default.entity-activation-range.monsters",
"proposed_value": "24",
"current_value": null
},
{
"id": "bukkit.limite-monstres",
"family": "bukkit",
"measure": "mobAI.mspt.avg",
"value": 445.8,
"unit": "ms",
"threshold": 50,
"control_test": "mobAI",
"file": "bukkit.yml",
"setting": "spawn-limits.monsters",
"proposed_value": "50",
"current_value": "70"
}
]
}Java start-up line
The options this server was started with, checked against its Java version. Corrections fix what is certainly wrong; leads to test are settings to measure, never applied for you.
Current line (27 arguments)
Values of system properties, agents and file paths are never shown: they appear as <redacted>.
-XX:ThreadPriorityPolicy=1 -XX:+UnlockExperimentalVMOptions -XX:+EnableJVMCIProduct -XX:+EnableJVMCI -XX:-UnlockExperimentalVMOptions -XX:+UseG1GC -XX:+ParallelRefProcEnabled -XX:MaxGCPauseMillis=200 -XX:+UnlockExperimentalVMOptions -XX:+DisableExplicitGC -XX:+AlwaysPreTouch -XX:G1NewSizePercent=30 -XX:G1MaxNewSizePercent=40 -XX:G1HeapRegionSize=8M -XX:G1ReservePercent=20 -XX:G1HeapWastePercent=5 -XX:G1MixedGCCountTarget=4 -XX:InitiatingHeapOccupancyPercent=15 -XX:G1MixedGCLiveThresholdPercent=90 -XX:G1RSetUpdatingPauseTimePercent=5 -XX:SurvivorRatio=32 -XX:+PerfDisableSharedMem -XX:MaxTenuringThreshold=1 -Dusing.aikars.flags=https://mcflags.emc.gs -Daikars.new.flags=true -Xmx11729M -Xms11729M
To correct
- To correctOption set more than once
-XX:+UnlockExperimentalVMOptionsThe JVM keeps only the last occurrence (-XX:+UnlockExperimentalVMOptions); the earlier ones do nothing but mislead whoever reads the line.
Corrected line to copy
- Removed
-XX:+UnlockExperimentalVMOptionsearlier occurrence; the JVM keeps -XX:+UnlockExperimentalVMOptions. - Removed
-XX:-UnlockExperimentalVMOptionsearlier occurrence; the JVM keeps -XX:+UnlockExperimentalVMOptions.
java -XX:ThreadPriorityPolicy=1 -XX:+EnableJVMCIProduct -XX:+EnableJVMCI -XX:+UseG1GC -XX:+ParallelRefProcEnabled -XX:MaxGCPauseMillis=200 -XX:+UnlockExperimentalVMOptions -XX:+DisableExplicitGC -XX:+AlwaysPreTouch -XX:G1NewSizePercent=30 -XX:G1MaxNewSizePercent=40 -XX:G1HeapRegionSize=8M -XX:G1ReservePercent=20 -XX:G1HeapWastePercent=5 -XX:G1MixedGCCountTarget=4 -XX:InitiatingHeapOccupancyPercent=15 -XX:G1MixedGCLiveThresholdPercent=90 -XX:G1RSetUpdatingPauseTimePercent=5 -XX:SurvivorRatio=32 -XX:+PerfDisableSharedMem -XX:MaxTenuringThreshold=1 -Xmx11729M -Xms11729M -jar server.jar noguiReplace server.jar with your server's jar. Paste the line where your server is started: the start-up parameters (or JVM arguments) of your hosting panel, or your start.sh / start.bat. Only the corrections above differ from your current line.
Your current line also carries 2 other arguments (system properties, agents): keep them between java and -jar. Their values are not shown here.
Re-run and compare
- Apply the corrected line where your server starts (panel start-up parameters, start.sh).
- Restart the server.
- Run a new benchmark:
/bench start - Compare the new report with this one.
This report is the before; add the new report to the comparison once it is in. A single run on each side says little: the gap between two runs of an unchanged server is noise.
Run conditions
What the plugin saw of the host during the benchmark, and what the site concludes from it. The plugin reports facts; the verdicts and their thresholds are the site's, the same ones the diagnosis and the API use.
The reference computation timed during the run kept its pace, and the hypervisor took no significant CPU time.
A CPU quota, a reduced CPU set or a memory limit applies to the server.
Contention during the run
- Reference computation slowdown
- 9.8 % (threshold 15 %)3 timings: 28.8 ms · 31.7 ms · 29.2 ms (a slow first timing alone does not count) — judged on the median of at least 40 samples per timing (plugin 2.6.2); host noise 7.9 %, the threshold widens to it, from 15 to 30 % — wait for the server to settle before Hopper, after chunk loading: 0.0 s (limit 30.0 s, after chunk generation) · longest wait, before Lighting Update: 0.4 s (limit 8.0 s, default limit)
- CPU time stolen by the hypervisor
- 0.2 % (threshold 5 %)
- Scheduling periods ended throttled by the quota
- 2 % (threshold 25 %)930.5 ms
- Time a task waited for a core
- 0.2 % (threshold 40 %)
Limits seen by the server
- CPU quota
- 8 cores
- CPU set
- 0-27 (28 cores)
- Memory limit
- 16 GiB
- Host logical CPUs
- 8
- Host memory
- 16 GiB
- Container signals
- cgroup-v2-namespace, overlay-root
Storage
- Filesystem
- ext4 · Local filesystem
- Direct 4K reads
- Honoured (1,019.5 µs against 1.7 µs from cache)
Single-core measurement
- Throughput (wall clock)
- 287.5M ops/s286.7M ops/s over the thread's CPU time
- Thread CPU time / wall time
- 1.003 (threshold 0.900)Below 1, the thread was not running for part of its slices (quota, neighbours, hypervisor steal when counted, JVM pauses). Close to 1 proves nothing.
- Spread from second to second
- 12.2 % (threshold 40 %)p10 276.3M · p50 297.1M · p90 312.6M ops/s over 9 seconds
- GC pauses during the measurement
- 0 ms (threshold 50 ms)0 collections, 0 ms of collection time
- Highest current CPU frequency
- not exposed by this systemadvertised maximum: 3,193 MHz
Witnesses of the single-core test (plugin 2.5.0). The score does not read this test, and none of these figures changes the run's stability verdict.
Host, memory and server state
- Time a task waited for storage
- 8.4 % (all tasks at once: 8.3 %)includes the disk test, which waits for storage by design: not judged for the run
- Time a task stalled on memory
- 0 % (all tasks at once: 0 %) (threshold 5 %)
- CPU quota during the run
- 8 coresno change observed
- Container memory
- peak 14,066 MiBlimit events: high 0 · max 0 · OOM kills 0
- Runnable tasks on the machine
- up to 10.8 on average during a test, for 8 logical CPUscontainer values (lxcfs): they say nothing about neighbours
- Server at the start of the run
- JVM up for 65.1 min3 benches before this oneoutside the bench world: 0 players · 289 chunks · 1 entities
Run pressures from plugin 2.4.0, the rest from plugin 2.6.0. None of these figures changes the score or the run's stability verdict; the diagnosis names the tests they concern.
Thresholds and their reasons are published in the benchmark methodology. Rules rc-7.
Every measurement
Each test in full. Tests in trouble are open; the others fold to their summary line.
Single-Core Stress
RedstoneUnstable19.34 TPS97.0 ms59.3s
Single-core stress via intensive redstone circuit updates.
- Heavy garbage collection9 collections, 155ms total
- The run did not hold steadyTPS P1=5.00<15, TPS StdDev=2.66>2, TPS min=5.00 (26% of avg)
Percentiles and stability
Raw data (72 fields)
Block PhysicsHealthy20.00 TPS29.0 ms24.8s
Single-core stress via falling-block physics.
- Heavy garbage collection3 collections, 47ms total
Percentiles and stability
Raw data (71 fields)
Exploration
Chunk LoadingCritical18.42 TPS12.5 ms21.9s
Chunk generation and loading speed across dispersed zones.
- memory_budget: no full GC observed by the pre-flight check, projected peak heap at 78.9%
- Heavy garbage collection2 collections, 501ms total
- The run did not hold steadyTPS avg=18.42<19, TPS P1=10.00<15, TPS StdDev=3.50>2, TPS min=10.00 (54% of avg)
Percentiles and stability
Raw data (61 fields)
Chunk TickingUnstable19.61 TPS39.7 ms41.4s
Server performance when ticking many loaded chunks with random tick updates.
- Heavy garbage collection2 collections, 120ms total
- The run did not hold steadyTPS P1=15.00<15
Percentiles and stability
Raw data (66 fields)
Lighting UpdateHealthy20.00 TPS3.3 ms25.0s
Lighting recomputation throughput on block place / break.
Percentiles and stability
Raw data (57 fields)
Entities
Mob AICritical5.09 TPS445.8 ms31.2s
Complex mob AI behaviours (villagers, iron golems, hostile mobs).
- Heavy garbage collection1 collections, 196ms total
- The run did not hold steadyTPS avg=5.09<19, TPS P1=5.00<15
Percentiles and stability
Raw data (77 fields)
Mechanics
HopperUnstable19.61 TPS4.4 ms19.4s
Intensive hopper chain item transfers.
- The run did not hold steadyTPS P1=12.50<15, TPS min=12.50 (64% of avg)
Percentiles and stability
Raw data (69 fields)
ExplosionUnstable19.81 TPS6.7 ms17.0s
Massive TNT explosions and block destruction.
- Heavy garbage collection1 collections, 210ms total
- The run did not hold steadyTPS P1=10.00<15, TPS min=10.00 (50% of avg)
Percentiles and stability
Raw data (115 fields)
Ticking Tile EntitiesHealthy20.00 TPS7.3 ms36.2s
Many ticking tile entities (furnaces, hoppers, etc.).
Percentiles and stability
Raw data (58 fields)
World SaveHealthy20.00 TPS0.8 ms4.8s
World save speed and TPS impact during save operations.
Percentiles and stability
Raw data (56 fields)
Bone Meal GrowthHealthy20.00 TPS1.5 ms23.3s
Rapid crop and tree growth via bone meal.
Percentiles and stability
Raw data (60 fields)
CPU
Single-Core (tick-aligned)Healthy20.00 TPS2.9 ms10.0s
Single-thread compute throughput measured on the main thread under the tick scheduler.
Percentiles and stability
Raw data (61 fields)
Multi-Core CPUHealthy20.00 TPS0.9 ms22.5s
Parallel computation throughput across worker threads.
Percentiles and stability
Raw data (56 fields)
Memory & Storage
Memory PerformanceHealthy20.00 TPS0.9 ms58.8s
RAM bandwidth and latency.
JVM measurements — do not reflect actual bare-metal performance.
Percentiles and stability
Raw data (56 fields)
Disk PerformanceCritical20.00 TPS1.1 ms409.7s
Sequential and random 4K read/write performance.
JVM measurements — do not reflect actual bare-metal performance.
6 phases stopped at their time budget before finishing: slow storage, so throughput and IOPS are computed over the time allowed rather than over the planned amount.
Percentiles and stability
Raw data (56 fields)
Network
Network (NBT)Critical10.62 TPS127.7 ms43.0s
Serialisation throughput — proxy for the rate at which the server can ship world data to clients.
- Heavy garbage collection5 collections, 219ms total
- The run did not hold steadyTPS avg=10.62<19, TPS P1=5.00<15, TPS StdDev=3.94>2, TPS min=5.00 (47% of avg)
Percentiles and stability
Raw data (60 fields)
›Machine and serverIntel(R) Xeon(R) CPU E5-2667 v3 @ 3.20GHz · Paper · 26.3
Score Breakdown
Detailed sub-category scores