This is the full report of the run “Stage 1 of the latency programme: five header-delay arms on a ten-region fleet”. Back to the run page.
r20261006-2058 · Fleet c3-hdb · c3-highcpu-8

Telcoin Network benchmark: mixed workload

Throughput is measured from blocks the chain produced (own senders vs every sender), resource use from Cloud Monitoring, Prometheus and the on-node sampler. Summaries cover the mixed phase. Hover any number for the JSON field it came from.

10 blaster(s), 6K senders, chain 32286, schema tn-blast-combined/v2 · mixed phase 2026-10-07 00:19–00:27 UTC · 1 observers, 10 validators

Preflight: WTEL from genesis (0x0000…37e1), v3 fee tier 3000, v4 available, Multicall3 present · Cooldown ended: all_landed ×1, idle_30s ×9

Headline numbers

Sustained mixed TPS (own)
6,623.1 tx/s
best 60-s mean of our own landed tx in the mixed phase
Sustained mixed TPS (chain)
6,623.1 tx/s
all tx in blocks, same window
Landed / accepted
98.5 %
3.5M of 3.56M accepted
Reverted (own)
0
0.00 % of landed
Latency p50
3,173 ms
submit → block timestamp; seen p50 7,821 ms
Latency p95
7,946 ms
seen p95 15,566 ms
Gas / s (own)
249M gas/s
chain 249M gas/s
Capacity at mix
9,328.4 tx/s
gas-limit basis 6,551.8 tx/s; block gas limit 30M

Per transaction type

Landed counts only our own senders; latency is submit → block timestamp.

TypeSubmittedAcceptedLandedRevertedUnlandedSkipsGas limitGas used avgp50p95
native_transfer1.35M1.35M1.34M013.2K024.1K21K3,158 ms7,942 ms
batched_self_send460K460K451K08.88K024.1K21K3,218 ms7,983 ms
uniswap_v2_swap412K412K405K07.69K0180K106K3,160 ms7,931 ms
stablecoin_transfer343K343K337K06.13K075K43.9K3,156 ms7,926 ms
wtel_wrap276K276K271K05.03K050K27.9K3,155 ms7,929 ms
wtel_unwrap184K184K180K03.37K045K35.2K3,160 ms7,914 ms
stablecoin_approve139K139K136K02.56K060K31.8K3,164 ms7,913 ms
uniswap_v2_add_liquidity91.7K91.7K89.9K01.77K0250K128K3,162 ms7,939 ms
faucet_drip72K72K70.4K01.66K42.2K150K107K3,351 ms11,160 ms
contract_deploy70.6K70.6K69.3K01.29K0140K74.2K3,160 ms7,921 ms
uniswap_v4_swap68.5K68.5K67.3K01.25K0250K124K3,168 ms7,929 ms
uniswap_v2_remove_liquidity45.8K45.8K45K08350220K129K3,160 ms7,908 ms
uniswap_v3_swap45.6K45.6K44.8K08170220K120K3,156 ms7,931 ms

Submission by RPC node

NodeSubmittedAcceptedAccepted %Top errorsSend p50Send p95
bench-validator-01
bench-validator-01:8545
626K626K100.0 %none8 ms29 ms
bench-validator-02
bench-validator-02:8545
61.1K44.6K72.9 %known 2.34K12 ms45 ms
bench-validator-03
bench-validator-03:8545
415K411K99.0 %known 2.38K3 ms18 ms
bench-validator-04
bench-validator-04:8545
508K502K98.8 %known 3.36K9 ms41 ms
bench-validator-05
bench-validator-05:8545
343K335K97.7 %known 4.13K5 ms25 ms
bench-validator-06
bench-validator-06:8545
141K118K83.8 %known 5.16K10 ms48 ms
bench-validator-07
bench-validator-07:8545
534K529K99.1 %known 2.7K5 ms25 ms
bench-validator-08
bench-validator-08:8545
602K599K99.6 %known 1.05K7 ms29 ms
bench-validator-09
bench-validator-09:8545
349K340K97.5 %known 4.65K6 ms25 ms
bench-validator-10
bench-validator-10:8545
126K117K92.2 %known 3.98K10 ms41 ms

Throughput over time

Shaded: phases outside the mixed window. Hairlines: epoch closes. Dashed orange: chaos events.

Landed transactions per second

own senders vs chain total, per block-timestamp second

Gas per second

own vs chain gas used, per block-timestamp second

CPU and memory per node

CPU utilization

compute.googleapis.com/instance/cpu/utilization, 60-s mean, % of all vCPUs; summaries over the mixed phase

Table view
NodeClassMeanp95Peak 60-s meanMax
bench-observer-11observer14.8 %16.9 %16.9 %16.9 %
bench-validator-01validator28.4 %31.6 %31.6 %31.6 %
bench-validator-02validator26.0 %29.7 %29.7 %29.7 %
bench-validator-03validator27.3 %30.7 %30.7 %30.7 %
bench-validator-04validator27.3 %30.9 %30.9 %30.9 %
bench-validator-05validator26.5 %30.7 %30.7 %30.7 %
bench-validator-06validator27.3 %30.3 %30.3 %30.3 %
bench-validator-07validator27.3 %30.9 %30.9 %30.9 %
bench-validator-08validator28.1 %31.5 %31.5 %31.5 %
bench-validator-09validator25.9 %29.4 %29.4 %29.4 %
bench-validator-10validator25.8 %29.8 %29.8 %29.8 %

CPU: mean vs peak 60-s mean

One row per node, over the mixed phase.

telcoin resident memory

agent processes/rss_usage summed over telcoin pids; summaries over the mixed phase

Table view
NodeClassMeanp95Peak 60-s meanMax
bench-observer-11observer3.07 GiB3.37 GiB3.37 GiB3.37 GiB
bench-validator-01validator4.86 GiB5.49 GiB5.49 GiB5.49 GiB
bench-validator-02validator5.47 GiB6.30 GiB6.30 GiB6.30 GiB
bench-validator-03validator5.06 GiB5.69 GiB5.69 GiB5.69 GiB
bench-validator-04validator5.23 GiB5.84 GiB5.84 GiB5.84 GiB
bench-validator-05validator5.00 GiB5.65 GiB5.65 GiB5.65 GiB
bench-validator-06validator5.58 GiB6.27 GiB6.27 GiB6.27 GiB
bench-validator-07validator5.09 GiB5.74 GiB5.74 GiB5.74 GiB
bench-validator-08validator5.15 GiB5.76 GiB5.76 GiB5.76 GiB
bench-validator-09validator5.06 GiB5.54 GiB5.54 GiB5.54 GiB
bench-validator-10validator5.56 GiB6.35 GiB6.35 GiB6.35 GiB

telcoin RSS: mean vs peak 60-s mean

One row per node, over the mixed phase.

Memory used

agent memory/percent_used, state=used; summaries over the mixed phase

Table view
NodeClassMeanp95Peak 60-s meanMax
bench-observer-11observer8.7 %9.0 %9.0 %9.0 %
bench-validator-01validator19.5 %21.5 %21.5 %21.5 %
bench-validator-02validator23.0 %26.4 %26.4 %26.4 %
bench-validator-03validator20.5 %22.6 %22.6 %22.6 %
bench-validator-04validator21.4 %23.4 %23.4 %23.4 %
bench-validator-05validator20.3 %22.5 %22.5 %22.5 %
bench-validator-06validator23.8 %26.3 %26.3 %26.3 %
bench-validator-07validator20.6 %22.8 %22.8 %22.8 %
bench-validator-08validator20.9 %22.9 %22.9 %22.9 %
bench-validator-09validator20.7 %22.6 %22.6 %22.6 %
bench-validator-10validator23.6 %26.8 %26.8 %26.8 %

Memory used: mean vs peak 60-s mean

One row per node, over the mixed phase.

Disk

Disk throttling (validator)
not collected –
metric type unavailable in project
Disk performance degraded (validator)
none –
any sample

Disk read + write throughput

instance/disk/{read,write}_bytes_count rate, all devices; summaries over the mixed phase

Table view
NodeClassMeanp95Peak 60-s meanMaxLimit
bench-observer-11observer82.9 MiB/s93.9 MiB/s93.9 MiB/s93.9 MiB/s800.0 MiB/s
bench-validator-01validator275.2 MiB/s299.2 MiB/s299.2 MiB/s299.2 MiB/s800.0 MiB/s
bench-validator-02validator272.8 MiB/s295.7 MiB/s295.7 MiB/s295.7 MiB/s800.0 MiB/s
bench-validator-03validator273.4 MiB/s298.5 MiB/s298.5 MiB/s298.5 MiB/s800.0 MiB/s
bench-validator-04validator274.8 MiB/s300.3 MiB/s300.3 MiB/s300.3 MiB/s800.0 MiB/s
bench-validator-05validator274.3 MiB/s297.4 MiB/s297.4 MiB/s297.4 MiB/s800.0 MiB/s
bench-validator-06validator272.7 MiB/s301.2 MiB/s301.2 MiB/s301.2 MiB/s800.0 MiB/s
bench-validator-07validator273.3 MiB/s298.3 MiB/s298.3 MiB/s298.3 MiB/s800.0 MiB/s
bench-validator-08validator274.1 MiB/s299.6 MiB/s299.6 MiB/s299.6 MiB/s800.0 MiB/s
bench-validator-09validator272.3 MiB/s295.6 MiB/s295.6 MiB/s295.6 MiB/s800.0 MiB/s
bench-validator-10validator271.8 MiB/s291.7 MiB/s291.7 MiB/s291.7 MiB/s800.0 MiB/s

Disk read + write IOPS

instance/disk/{read,write}_ops_count rate, all devices; summaries over the mixed phase

Table view
NodeClassMeanp95Peak 60-s meanMaxLimit
bench-observer-11observer17.8K IOPS20.1K IOPS20.1K IOPS20.1K IOPS20K IOPS
bench-validator-01validator17.5K IOPS20K IOPS20K IOPS20K IOPS20K IOPS
bench-validator-02validator17.5K IOPS20K IOPS20K IOPS20K IOPS20K IOPS
bench-validator-03validator17.4K IOPS19.9K IOPS19.9K IOPS19.9K IOPS20K IOPS
bench-validator-04validator17.4K IOPS20.1K IOPS20.1K IOPS20.1K IOPS20K IOPS
bench-validator-05validator17.4K IOPS19.9K IOPS19.9K IOPS19.9K IOPS20K IOPS
bench-validator-06validator17.5K IOPS20K IOPS20K IOPS20K IOPS20K IOPS
bench-validator-07validator17.4K IOPS20K IOPS20K IOPS20K IOPS20K IOPS
bench-validator-08validator17.4K IOPS20K IOPS20K IOPS20K IOPS20K IOPS
bench-validator-09validator17.3K IOPS19.9K IOPS19.9K IOPS19.9K IOPS20K IOPS
bench-validator-10validator17.4K IOPS20.1K IOPS20.1K IOPS20.1K IOPS20K IOPS
Disk limits used
NodeMachinevCPUDiskIOPS limitMiB/s limitBasis
bench-observer-11c3-highcpu-88hyperdisk-balanced 100 GB20K IOPS800.0 MiB/shyperdisk-balanced 100 GB: provisioned 20,000 IOPS, provisioned 800 MiB/s; c3 8 vCPU cap 50,000 IOPS / 800 MiB/s; reads and writes share this budget
bench-validator-01c3-highcpu-88hyperdisk-balanced 100 GB20K IOPS800.0 MiB/shyperdisk-balanced 100 GB: provisioned 20,000 IOPS, provisioned 800 MiB/s; c3 8 vCPU cap 50,000 IOPS / 800 MiB/s; reads and writes share this budget
bench-validator-02c3-highcpu-88hyperdisk-balanced 100 GB20K IOPS800.0 MiB/shyperdisk-balanced 100 GB: provisioned 20,000 IOPS, provisioned 800 MiB/s; c3 8 vCPU cap 50,000 IOPS / 800 MiB/s; reads and writes share this budget
bench-validator-03c3-highcpu-88hyperdisk-balanced 100 GB20K IOPS800.0 MiB/shyperdisk-balanced 100 GB: provisioned 20,000 IOPS, provisioned 800 MiB/s; c3 8 vCPU cap 50,000 IOPS / 800 MiB/s; reads and writes share this budget
bench-validator-04c3-highcpu-88hyperdisk-balanced 100 GB20K IOPS800.0 MiB/shyperdisk-balanced 100 GB: provisioned 20,000 IOPS, provisioned 800 MiB/s; c3 8 vCPU cap 50,000 IOPS / 800 MiB/s; reads and writes share this budget
bench-validator-05c3-highcpu-88hyperdisk-balanced 100 GB20K IOPS800.0 MiB/shyperdisk-balanced 100 GB: provisioned 20,000 IOPS, provisioned 800 MiB/s; c3 8 vCPU cap 50,000 IOPS / 800 MiB/s; reads and writes share this budget
bench-validator-06c3-highcpu-88hyperdisk-balanced 100 GB20K IOPS800.0 MiB/shyperdisk-balanced 100 GB: provisioned 20,000 IOPS, provisioned 800 MiB/s; c3 8 vCPU cap 50,000 IOPS / 800 MiB/s; reads and writes share this budget
bench-validator-07c3-highcpu-88hyperdisk-balanced 100 GB20K IOPS800.0 MiB/shyperdisk-balanced 100 GB: provisioned 20,000 IOPS, provisioned 800 MiB/s; c3 8 vCPU cap 50,000 IOPS / 800 MiB/s; reads and writes share this budget
bench-validator-08c3-highcpu-88hyperdisk-balanced 100 GB20K IOPS800.0 MiB/shyperdisk-balanced 100 GB: provisioned 20,000 IOPS, provisioned 800 MiB/s; c3 8 vCPU cap 50,000 IOPS / 800 MiB/s; reads and writes share this budget
bench-validator-09c3-highcpu-88hyperdisk-balanced 100 GB20K IOPS800.0 MiB/shyperdisk-balanced 100 GB: provisioned 20,000 IOPS, provisioned 800 MiB/s; c3 8 vCPU cap 50,000 IOPS / 800 MiB/s; reads and writes share this budget
bench-validator-10c3-highcpu-88hyperdisk-balanced 100 GB20K IOPS800.0 MiB/shyperdisk-balanced 100 GB: provisioned 20,000 IOPS, provisioned 800 MiB/s; c3 8 vCPU cap 50,000 IOPS / 800 MiB/s; reads and writes share this budget

Network

Network sent

instance/network/sent_bytes_count rate; summaries over the mixed phase

Table view
NodeClassMeanp95Peak 60-s meanMax
bench-observer-11observer1.2 Mbps1.3 Mbps1.3 Mbps1.3 Mbps
bench-validator-01validator104.4 Mbps116.9 Mbps116.9 Mbps116.9 Mbps
bench-validator-02validator90.5 Mbps103.3 Mbps103.3 Mbps103.3 Mbps
bench-validator-03validator94.9 Mbps105.0 Mbps105.0 Mbps105.0 Mbps
bench-validator-04validator96.1 Mbps110.4 Mbps110.4 Mbps110.4 Mbps
bench-validator-05validator93.7 Mbps108.6 Mbps108.6 Mbps108.6 Mbps
bench-validator-06validator91.3 Mbps100.2 Mbps100.2 Mbps100.2 Mbps
bench-validator-07validator97.1 Mbps109.9 Mbps109.9 Mbps109.9 Mbps
bench-validator-08validator99.0 Mbps111.0 Mbps111.0 Mbps111.0 Mbps
bench-validator-09validator92.5 Mbps104.2 Mbps104.2 Mbps104.2 Mbps
bench-validator-10validator90.3 Mbps103.7 Mbps103.7 Mbps103.7 Mbps

Network received

instance/network/received_bytes_count rate; summaries over the mixed phase

Table view
NodeClassMeanp95Peak 60-s meanMax
bench-observer-11observer6.6 Mbps7.5 Mbps7.5 Mbps7.5 Mbps
bench-validator-01validator10.8 Mbps12.0 Mbps12.0 Mbps12.0 Mbps
bench-validator-02validator6.9 Mbps9.2 Mbps9.2 Mbps9.2 Mbps
bench-validator-03validator9.0 Mbps10.3 Mbps10.3 Mbps10.3 Mbps
bench-validator-04validator9.8 Mbps11.7 Mbps11.7 Mbps11.7 Mbps
bench-validator-05validator8.5 Mbps10.8 Mbps10.8 Mbps10.8 Mbps
bench-validator-06validator8.2 Mbps10.2 Mbps10.2 Mbps10.2 Mbps
bench-validator-07validator10.1 Mbps11.5 Mbps11.5 Mbps11.5 Mbps
bench-validator-08validator10.7 Mbps12.3 Mbps12.3 Mbps12.3 Mbps
bench-validator-09validator8.1 Mbps10.2 Mbps10.2 Mbps10.2 Mbps
bench-validator-10validator6.9 Mbps8.8 Mbps8.8 Mbps8.8 Mbps

Consensus health

MetricQueryClassNodesMedian of node meansWorst peak 60-s meanMax
tn_primary_commit_latency_secondsp95validator102,411 ms5,258 ms6,250 ms
tn_engine_execution_duration_secondsp95validator102,591 ms4,154 ms4,450 ms
tn_engine_execution_duration_secondsp95observer12,268 ms2,487 ms2,500 ms
tn_worker_batch_size_bytesp95validator10241.1 KiB246.4 KiB246.4 KiB
tn_worker_batch_transactionsp95validator104861K1K
tn_executor_output_transactionsp95validator101K1K1K
tn_primary_consensus_output_bytesmaxvalidator10662.8 KiB2.2 MiB2.2 MiB
tn_primary_consensus_output_bytesmaxobserver1659.1 KiB1.6 MiB1.6 MiB
tn_engine_queued_outputsmaxvalidator101.383.54
tn_engine_queued_outputsmaxobserver10.9382.53
tn_primary_roundmaxvalidator101.57K1.96K2K
tn_primary_roundmaxobserver11.52K1.94K1.97K
tn_primary_committed_roundmaxvalidator101.57K1.96K2K
tn_primary_committed_roundmaxobserver11.52K1.94K1.97K
tn_epoch_currentmaxvalidator108.098.759
tn_epoch_currentmaxobserver18.098.759
tn_batch_builder_pending_pool_transactionsmaxvalidator101.03K6K6.41K
tn_batch_builder_pending_pool_transactionsmaxobserver1000
reth_process_resident_memory_bytesmaxvalidator105.22 GiB6.33 GiB6.35 GiB
reth_process_resident_memory_bytesmaxobserver13.09 GiB3.35 GiB3.37 GiB
tn_primary_subdags_committed_totalratevalidator100.710.8670.911
tn_primary_subdags_committed_totalrateobserver1000
tn_primary_round_laglagvalidator102.3147
tn_primary_round_laglagobserver1000
tn_batch_builder_seal_failures_totalratevalidator1000
tn_node_modemaxvalidator10000
tn_node_modemaxobserver1000

Commit latency p95

tn_primary_commit_latency_seconds

Round lag

tn_primary_round − tn_primary_committed_round

Engine queued outputs

tn_engine_queued_outputs, max by node

On-node sampler

NodeClassPeak RssAnonPeak RssFilePeak cachePSI memPSI ioPSI cpudb growthstatic_files growthRoot usedOOM Δ
bench-observer-11observer460 MiB2.93 GiB64 MiB 6.3 %0.0 %9.6 %0.1 %0 B3.67 GiB14.0 %0
bench-validator-01validator2.49 GiB3.01 GiB64 MiB 6.3 %0.0 %11.8 %2.8 %0 B3.67 GiB15.0 %0
bench-validator-02validator3.31 GiB3.00 GiB64 MiB 6.3 %0.0 %13.3 %2.7 %0 B3.67 GiB14.0 %0
bench-validator-03validator2.68 GiB3.01 GiB64 MiB 6.3 %0.0 %13.8 %2.4 %0 B3.67 GiB15.0 %0
bench-validator-04validator2.84 GiB3.01 GiB64 MiB 6.3 %0.0 %13.8 %2.6 %0 B3.67 GiB15.0 %0
bench-validator-05validator2.65 GiB3.01 GiB64 MiB 6.3 %0.0 %12.3 %2.5 %0 B3.67 GiB15.0 %0
bench-validator-06validator3.30 GiB2.97 GiB64 MiB 6.3 %0.0 %12.6 %2.6 %0 B3.67 GiB15.0 %0
bench-validator-07validator2.75 GiB3.01 GiB64 MiB 6.3 %0.0 %13.8 %2.6 %0 B3.67 GiB15.0 %0
bench-validator-08validator2.76 GiB3.01 GiB64 MiB 6.3 %0.0 %12.3 %2.7 %0 B3.67 GiB15.0 %0
bench-validator-09validator2.68 GiB2.87 GiB64 MiB 6.3 %0.0 %11.9 %2.2 %0 B3.67 GiB15.0 %0
bench-validator-10validator3.38 GiB2.98 GiB64 MiB 6.3 %0.0 %12.8 %2.4 %0 B3.67 GiB15.0 %0

RssAnon vs RssFile

/proc/<pid>/status of the telcoin process on the node with the largest RSS

consensus-db/cache size

du -sb of consensus-db/cache vs the 1 GiB ceiling

Pressure stall (some, avg10)

/proc/pressure on the node with the highest memory PSI

Headroom

Per role and per node, from sizing.json (p95 over the landed window; PSI and RssAnon max over the whole sampler run). The lines follow the hardware doc's tier rules: CPU p95 over 70 % of vCPUs; memory used p95 over 75 % of RAM, or telcoin RSS p95 over 85 % of RAM; disk write p95 over 60 % of the PD MiB/s limit, or write IOPS p95 over 60 % of the IOPS limit; IO pressure over 20 % or memory pressure over 5 % (PSI some avg10, max); network p95 over 50 % of 1 Gbps; root disk over 80 %. A line crossed on any node puts the role at the limit; a nonzero reading within 15 points of its line is tight; otherwise the role has headroom. OOM kills, disk throttling and degraded disk performance are flagged too.

SignalClassWorst nodeValueShare of capacityLineNodes over / tightStatus
CPU p95validatorbench-validator-0635.0 %35.0 %> 70 %0 / 0 headroom
Memory used p95validatorbench-validator-1026.8 %26.8 %> 75 %0 / 0 headroom
telcoin RSS p95validatorbench-validator-106.82 GB39.7 %> 85 %0 / 0 headroom
PSI memory maxvalidatorbench-validator-010.0 %0.0 %> 5 %0 / 0 headroom
RssAnon peakvalidatorbench-validator-074.45 GB25.9 %info– info
Disk write p95validatorbench-validator-07332 MB/s39.6 %> 60 %0 / 0 headroom
Write IOPS p95validatorbench-validator-0420,105100.5 %> 60 %10 / 0 at the limit
PSI io maxvalidatorbench-validator-0222.5 %22.5 %> 20 %8 / 2 at the limit
Root disk maxvalidatorbench-validator-0914.6 %14.6 %> 80 %0 / 0 headroom
Network p95validatorbench-validator-01117 Mbps11.7 %> 50 %0 / 0 headroom
Consensus cache peakvalidatorbench-validator-010.06 GiB6.3 %info– info
OOM killsvalidator––anyany– none
Disk throttlingvalidator––metric type unavailable in projectany– not collected
Disk performance degradedvalidator––any sampleany– none
CPU p95observerbench-observer-1117.2 %17.2 %> 70 %0 / 0 headroom
Memory used p95observerbench-observer-119.0 %9.0 %> 75 %0 / 0 headroom
telcoin RSS p95observerbench-observer-113.62 GB21.1 %> 85 %0 / 0 headroom
PSI memory maxobserverbench-observer-110.0 %0.0 %> 5 %0 / 0 headroom
RssAnon peakobserverbench-observer-110.58 GB3.4 %info– info
Disk write p95observerbench-observer-1198 MB/s11.7 %> 60 %0 / 0 headroom
Write IOPS p95observerbench-observer-1120,078100.4 %> 60 %1 / 0 at the limit
PSI io maxobserverbench-observer-1112.9 %12.9 %> 20 %0 / 1 tight
Root disk maxobserverbench-observer-1113.2 %13.2 %> 80 %0 / 0 headroom
Network p95observerbench-observer-1110 Mbps1.0 %> 50 %0 / 0 headroom
Consensus cache peakobserverbench-observer-110.06 GiB6.3 %info– info
OOM killsobserver––anyany– none
Disk throttlingobserver––metric type unavailable in projectany– not collected
Disk performance degradedobserver––any sampleany– none

Per node

NodeClassCPU p95Memory used p95telcoin RSS p95PSI memory maxRssAnon peakDisk write p95Write IOPS p95PSI io maxRoot disk maxNetwork p95Consensus cache peak
Line> 70 %> 75 %> 85 %> 5 %info> 60 %> 60 %> 20 %> 80 %> 50 %info
bench-validator-01validator32.7 %21.5 %5.90 GB 34 %0.0 %3.68 GB 21 %319 MB/s 38 %20,025 100 %20.8 %14.0 %117 Mbps 12 %0.06 GiB 6 %
bench-validator-02validator32.3 %26.4 %6.76 GB 39 %0.0 %3.72 GB 22 %324 MB/s 39 %20,009 100 %22.5 %14.0 %103 Mbps 10 %0.06 GiB 6 %
bench-validator-03validator31.6 %22.6 %6.10 GB 36 %0.0 %3.57 GB 21 %325 MB/s 39 %19,943 100 %21.2 %14.1 %105 Mbps 11 %0.06 GiB 6 %
bench-validator-04validator33.1 %23.4 %6.27 GB 36 %0.0 %3.22 GB 19 %328 MB/s 39 %20,105 101 %20.9 %14.1 %110 Mbps 11 %0.06 GiB 6 %
bench-validator-05validator31.8 %22.5 %6.06 GB 35 %0.0 %3.38 GB 20 %327 MB/s 39 %19,908 100 %20.9 %14.1 %109 Mbps 11 %0.06 GiB 6 %
bench-validator-06validator35.0 %26.3 %6.74 GB 39 %0.0 %4.29 GB 25 %321 MB/s 38 %19,972 100 %20.7 %14.1 %100 Mbps 10 %0.06 GiB 6 %
bench-validator-07validator32.1 %22.8 %6.17 GB 36 %0.0 %4.45 GB 26 %332 MB/s 40 %19,976 100 %20.2 %14.2 %110 Mbps 11 %0.06 GiB 6 %
bench-validator-08validator31.7 %22.9 %6.18 GB 36 %0.0 %3.46 GB 20 %328 MB/s 39 %19,971 100 %19.9 %14.2 %111 Mbps 11 %0.06 GiB 6 %
bench-validator-09validator32.5 %22.6 %5.95 GB 35 %0.0 %3.47 GB 20 %332 MB/s 40 %19,937 100 %19.4 %14.6 %107 Mbps 11 %0.06 GiB 6 %
bench-validator-10validator31.6 %26.8 %6.82 GB 40 %0.0 %4.02 GB 23 %332 MB/s 40 %20,059 100 %21.2 %14.2 %104 Mbps 10 %0.06 GiB 6 %
bench-observer-11observer17.2 %9.0 %3.62 GB 21 %0.0 %0.58 GB 3 %98 MB/s 12 %20,078 100 %12.9 %13.2 %10 Mbps 1 %0.06 GiB 6 %

Small figures are the share of capacity: RAM for RSS, the PD limit for disk write (VM cap applied), 1 Gbps for network. Hover a cell for its JSON path.

Verdict

Fleet A vs Fleet B

A = Fleet c3-hdb · c3-highcpu-8 · r20261006-2058
B = Fleet c3-hdb · c3-highcpu-8 · r20261006-0458

Verdict

at the limit c3-highcpu-8 validators at 6,623 sustained TPS: at the limit on disk. Disk at the limit: write IOPS p95 19,908–20,105, up to 100.5 % of the 20,000 IOPS limit, over the 60 % line on all 10 validators (bench-validator-01, 02, 03, 04, 05, 06, 07, 08, 09, 10); IO pressure (PSI some avg10) max 20.2–22.5 %, over the 20 % line on 8 of 10 validators (bench-validator-01, 02, 03, 04, 05, 06, 07, 10). CPU has headroom: CPU p95 35.0 % of 8 vCPUs (bench-validator-06). Memory has headroom: memory used p95 26.8 % of RAM (bench-validator-10); telcoin RSS p95 6.82 GB, 39.7 % of 16 GiB RAM (bench-validator-10); memory pressure (PSI some avg10) max 0.0 % (bench-validator-01). Network has headroom: network p95 117 Mbps, 11.7 % of 1 Gbps (bench-validator-01). Next step: a faster volume (pd-ssd adds 30 IOPS and 0.48 MiB/s per GB up to the per-VM cap; past the cap, a larger VM).

at the limit c3-highcpu-8 observers at 6,623 sustained TPS: at the limit on disk. Disk at the limit: write IOPS p95 20,078, 100.4 % of the 20,000 IOPS limit on bench-observer-11, over the 60 % line; IO pressure (PSI some avg10) max 12.9 % (bench-observer-11), tight against the 20 % line. CPU has headroom: CPU p95 17.2 % of 8 vCPUs (bench-observer-11). Memory has headroom: memory used p95 9.0 % of RAM (bench-observer-11); telcoin RSS p95 3.62 GB, 21.1 % of 16 GiB RAM (bench-observer-11); memory pressure (PSI some avg10) max 0.0 % (bench-observer-11). Network has headroom: network p95 10 Mbps, 1.0 % of 1 Gbps (bench-observer-11). Next step: a faster volume (pd-ssd adds 30 IOPS and 0.48 MiB/s per GB up to the per-VM cap; past the cap, a larger VM).

at the limit c3-highcpu-8 validators at 5,953 sustained TPS: at the limit on disk. Disk at the limit: write p95 505–508 MB/s, up to 60.5 % of the 800 MiB/s limit, over the 60 % line on 2 of 10 validators (bench-validator-09, 10); IO pressure (PSI some avg10) max 26.4–32.2 %, over the 20 % line on all 10 validators (bench-validator-01, 02, 03, 04, 05, 06, 07, 08, 09, 10). Memory tight: memory pressure (PSI some avg10) max 0.0 % (bench-validator-03), tight against the 5 % line. CPU has headroom: CPU p95 31.3 % of 8 vCPUs (bench-validator-02). Network has headroom: network p95 103 Mbps, 10.3 % of 1 Gbps (bench-validator-06). Next step: a faster volume (pd-ssd adds 30 IOPS and 0.48 MiB/s per GB up to the per-VM cap; past the cap, a larger VM).

at the limit c3-highcpu-8 observers at 5,953 sustained TPS: at the limit on disk. Disk at the limit: IO pressure (PSI some avg10) max 29.3 % on bench-observer-11, over the 20 % line; write p95 475 MB/s, 56.6 % of the 800 MiB/s limit (bench-observer-11), tight against the 60 % line. CPU has headroom: CPU p95 20.3 % of 8 vCPUs (bench-observer-11). Memory has headroom: memory used p95 11.9 % of RAM (bench-observer-11); telcoin RSS p95 4.38 GB, 25.5 % of 16 GiB RAM (bench-observer-11); memory pressure (PSI some avg10) max 0.0 % (bench-observer-11). Network has headroom: network p95 9 Mbps, 0.9 % of 1 Gbps (bench-observer-11). Next step: a faster volume (pd-ssd adds 30 IOPS and 0.48 MiB/s per GB up to the per-VM cap; past the cap, a larger VM).

Validators: Fleet A (c3-highcpu-8) is at the limit on disk, with headroom on CPU, memory and network; Fleet B (c3-highcpu-8) is at the limit on disk, tight on memory, with headroom on CPU and network. The same disk limit binds both fleets: every validator runs a 100 GB hyperdisk-balanced, which allows 20,000 IOPS on both; its throughput limit is 800 MiB/s on both. Disk is at the limit on both (Fleet A (c3-highcpu-8): write IOPS p95 20,105, 100.5 % of the 20,000 IOPS limit, IO pressure (PSI some avg10) max 22.5 %; Fleet B (c3-highcpu-8): write p95 508 MB/s, 60.5 % of the 800 MiB/s limit, IO pressure (PSI some avg10) max 32.2 %). Changing the machine shape did not clear it; the volume is the binding resource.

Observers: Fleet A (c3-highcpu-8) is at the limit on disk, with headroom on CPU, memory and network; Fleet B (c3-highcpu-8) is at the limit on disk, with headroom on CPU, memory and network. The same disk limit binds both fleets: every observer runs a 100 GB hyperdisk-balanced, which allows 20,000 IOPS on both; its throughput limit is 800 MiB/s on both. Disk is at the limit on both (Fleet A (c3-highcpu-8): write IOPS p95 20,078, 100.4 % of the 20,000 IOPS limit; Fleet B (c3-highcpu-8): IO pressure (PSI some avg10) max 29.3 %). Changing the machine shape did not clear it; the volume is the binding resource.

KPIFleet AFleet BB ÷ A
Sustained mixed TPS (own)6,623.1 tx/s5,952.9 tx/s0.90×
Sustained mixed TPS (chain)6,623.1 tx/s5,952.9 tx/s0.90×
Landed / accepted98.5 %99.6 %1.01×
Reverted (own)00–
Latency p503,173 ms3,383 ms1.07×
Latency p957,946 ms17,983 ms2.26×
Gas / s (own)249M gas/s167M gas/s0.67×
Capacity at mix9,328.4 tx/s6,852.7 tx/s0.73×

Headroom, worst node per class

SignalClassFleet AFleet BLine
CPU p95validator35.0 % headroom31.3 % headroom> 70 %
Memory used p95validator26.8 % headroom50.6 % headroom> 75 %
telcoin RSS p95validator6.82 GB 39.7 % headroom10.31 GB 60.0 % headroom> 85 %
PSI memory maxvalidator0.0 % headroom0.0 % tight> 5 %
Disk write p95validator332 MB/s 39.6 % headroom508 MB/s 60.5 % at the limit> 60 %
Write IOPS p95validator20,105 100.5 % at the limit5,324 26.6 % headroom> 60 %
PSI io maxvalidator22.5 % at the limit32.2 % at the limit> 20 %
Root disk maxvalidator14.6 % headroom13.6 % headroom> 80 %
Network p95validator117 Mbps 11.7 % headroom103 Mbps 10.3 % headroom> 50 %
CPU p95observer17.2 % headroom20.3 % headroom> 70 %
Memory used p95observer9.0 % headroom11.9 % headroom> 75 %
telcoin RSS p95observer3.62 GB 21.1 % headroom4.38 GB 25.5 % headroom> 85 %
PSI memory maxobserver0.0 % headroom0.0 % headroom> 5 %
Disk write p95observer98 MB/s 11.7 % headroom475 MB/s 56.6 % tight> 60 %
Write IOPS p95observer20,078 100.4 % at the limit2,101 10.5 % headroom> 60 %
PSI io maxobserver12.9 % tight29.3 % at the limit> 20 %
Root disk maxobserver13.2 % headroom14.3 % headroom> 80 %
Network p95observer10 Mbps 1.0 % headroom9 Mbps 0.9 % headroom> 50 %

Own TPS, trailing 60-s mean

aligned on each fleet's mixed-phase start

CPU, hottest validator

aligned on each fleet's mixed-phase start

Memory used, hottest validator

aligned on each fleet's mixed-phase start

telcoin RSS, hottest validator

aligned on each fleet's mixed-phase start

Disk MiB/s, hottest validator

aligned on each fleet's mixed-phase start

Network sent, hottest validator

aligned on each fleet's mixed-phase start

Reference chains

Not collected: reference chains.