Frankfurt DoubleZero TPU vs Public Internet
For each eligible leader run, the benchmark submitted one transaction through DoubleZero and one through Public Internet. Every performance delta below requires both transactions to be observed in the same Solana slot.
Overall same-slot comparison
Each delta is candidate minus baseline. Negative values favor the candidate. Percentiles select observed values without interpolation.
| Measure | Pairs | DoubleZero earlier | Same | Public Internet earlier | Delta p50 | Delta p95 | Delta p99 |
|---|---|---|---|---|---|---|---|
| First observation time | 201 | 115 | 0 | 86 | -0.006 ms | 0.018 ms | 4.408 ms |
| Processed transaction index | 201 | 115 | 0 | 86 | -2 | 4 | 20 |
| Entry index | 201 | 113 | 4 | 84 | -2 | 3 | 6 |
| Preceding PoH ticks | 201 | 91 | 106 | 4 | 0 | 0 | 1 |
Coverage outside the comparison population: 3 different-slot pairs, 2 baseline-only, 34 candidate-only, and 6 observed through neither route. These outcomes are reported as coverage and are never converted into timing or placement deltas.
By leader city
Cities and countries come from the getLeaderSlots snapshots saved during the run.
| Cohort | Validators | Same-slot n | Observation DoubleZero/same/Public Internet | Observation p50 / p95 / p99 | Tx index DoubleZero/same/Public Internet | Tx index p50 / p95 / p99 | Entry DoubleZero/same/Public Internet | Entry p50 / p95 / p99 | PoH DoubleZero/same/Public Internet | PoH p50 / p95 / p99 |
|---|---|---|---|---|---|---|---|---|---|---|
| Beauharnois, CA | 1 | 73 | 71 / 0 / 2 | -5.395 ms / -0.004 ms / 31.417 ms | 71 / 0 / 2 | -23 / -5 / 36 | 71 / 1 / 1 | -22 / -6 / 35 | 71 / 1 / 1 | -3 / -1 / 4 |
| London, GB | 2 | 56 | 0 / 0 / 56 | 0.009 ms / 0.112 ms / 4.408 ms | 0 / 0 / 56 | 1 / 4 / 7 | 0 / 0 / 56 | 1 / 4 / 7 | 0 / 56 / 0 | 0 / 0 / 0 |
| Luxembourg, LU | 2 | 28 | 17 / 0 / 11 | -0.005 ms / 0.023 ms / 7.263 ms | 17 / 0 / 11 | -1 / 5 / 5 | 17 / 0 / 11 | -1 / 4 / 5 | 4 / 24 / 0 | 0 / 0 / 0 |
| Amsterdam, NL | 5 | 12 | 6 / 0 / 6 | 0.005 ms / 0.013 ms / 0.013 ms | 6 / 0 / 6 | 1 / 7 / 7 | 4 / 3 / 5 | 0 / 5 / 5 | 0 / 11 / 1 | 0 / 1 / 1 |
| Aubervilliers, FR | 1 | 11 | 3 / 0 / 8 | 0.011 ms / 0.574 ms / 0.574 ms | 3 / 0 / 8 | 2 / 35 / 35 | 3 / 0 / 8 | 2 / 6 / 6 | 0 / 9 / 2 | 0 / 1 / 1 |
| Singapore, SG | 1 | 11 | 11 / 0 / 0 | -0.009 ms / -0.005 ms / -0.005 ms | 11 / 0 / 0 | -7 / -2 / -2 | 11 / 0 / 0 | -8 / -3 / -3 | 10 / 1 / 0 | -1 / 0 / 0 |
| São Paulo, BR | 2 | 2 | 2 / 0 / 0 | -17.509 ms / -17.509 ms / -17.509 ms | 2 / 0 / 0 | -40 / -40 / -40 | 2 / 0 / 0 | -45 / -45 / -45 | 2 / 0 / 0 | -5 / -5 / -5 |
| Tokyo, JP | 1 | 2 | 2 / 0 / 0 | -20.209 ms / -20.209 ms / -20.209 ms | 2 / 0 / 0 | -54 / -54 / -54 | 2 / 0 / 0 | -48 / -48 / -48 | 2 / 0 / 0 | -5 / -5 / -5 |
| Bloomfield, US | 1 | 1 | 0 / 0 / 1 | 0.011 ms / 0.011 ms / 0.011 ms | 0 / 0 / 1 | 2 / 2 / 2 | 0 / 0 / 1 | 2 / 2 / 2 | 0 / 1 / 0 | 0 / 0 / 0 |
| Hong Kong, HK | 1 | 1 | 0 / 0 / 1 | 0.005 ms / 0.005 ms / 0.005 ms | 0 / 0 / 1 | 1 / 1 / 1 | 0 / 0 / 1 | 1 / 1 / 1 | 0 / 1 / 0 | 0 / 0 / 0 |
| London, EN | 1 | 1 | 1 / 0 / 0 | -0.011 ms / -0.011 ms / -0.011 ms | 1 / 0 / 0 | -17 / -17 / -17 | 1 / 0 / 0 | -10 / -10 / -10 | 1 / 0 / 0 | -2 / -2 / -2 |
| Los Angeles, US | 1 | 1 | 1 / 0 / 0 | -0.009 ms / -0.009 ms / -0.009 ms | 1 / 0 / 0 | -3 / -3 / -3 | 1 / 0 / 0 | -4 / -4 / -4 | 1 / 0 / 0 | -1 / -1 / -1 |
| Prague, CZ | 1 | 1 | 0 / 0 / 1 | 0.011 ms / 0.011 ms / 0.011 ms | 0 / 0 / 1 | 1 / 1 / 1 | 0 / 0 / 1 | 1 / 1 / 1 | 0 / 1 / 0 | 0 / 0 / 0 |
| Stockholm, SE | 1 | 1 | 1 / 0 / 0 | -0.013 ms / -0.013 ms / -0.013 ms | 1 / 0 / 0 | -1 / -1 / -1 | 1 / 0 / 0 | -1 / -1 / -1 | 0 / 1 / 0 | 0 / 0 / 0 |
Outcome triplets are candidate earlier / same / baseline earlier. Every percentile uses only complete same-slot values for that measure.
By distance from sender
Distance is calculated from the saved validator coordinates to the sender coordinates supplied to the report command. Missing coordinates remain an explicit unavailable cohort.
| Cohort | Validators | Same-slot n | Observation DoubleZero/same/Public Internet | Observation p50 / p95 / p99 | Tx index DoubleZero/same/Public Internet | Tx index p50 / p95 / p99 | Entry DoubleZero/same/Public Internet | Entry p50 / p95 / p99 | PoH DoubleZero/same/Public Internet | PoH p50 / p95 / p99 |
|---|---|---|---|---|---|---|---|---|---|---|
| Under 1,000 km | 12 | 109 | 27 / 0 / 82 | 0.008 ms / 0.052 ms / 4.408 ms | 27 / 0 / 82 | 1 / 5 / 20 | 25 / 3 / 81 | 1 / 5 / 6 | 5 / 101 / 3 | 0 / 0 / 1 |
| At least 5,000 km | 7 | 90 | 86 / 0 / 4 | -0.312 ms / -0.004 ms / 31.417 ms | 86 / 0 / 4 | -22 / -2 / 36 | 86 / 1 / 3 | -19 / -3 / 35 | 85 / 4 / 1 | -2 / 0 / 4 |
| 1,000-5,000 km | 1 | 1 | 1 / 0 / 0 | -0.013 ms / -0.013 ms / -0.013 ms | 1 / 0 / 0 | -1 / -1 / -1 | 1 / 0 / 0 | -1 / -1 / -1 | 0 / 1 / 0 | 0 / 0 / 0 |
| Coordinate unavailable | 1 | 1 | 1 / 0 / 0 | -0.009 ms / -0.009 ms / -0.009 ms | 1 / 0 / 0 | -3 / -3 / -3 | 1 / 0 / 0 | -4 / -4 / -4 | 1 / 0 / 0 | -1 / -1 / -1 |
Outcome triplets are candidate earlier / same / baseline earlier. Every percentile uses only complete same-slot values for that measure.
By validator identity
Each validator remains separate. Repeated leader runs from one validator are not independent validator samples; small cohorts are descriptive only.
| Cohort | Validators | Same-slot n | Observation DoubleZero/same/Public Internet | Observation p50 / p95 / p99 | Tx index DoubleZero/same/Public Internet | Tx index p50 / p95 / p99 | Entry DoubleZero/same/Public Internet | Entry p50 / p95 / p99 | PoH DoubleZero/same/Public Internet | PoH p50 / p95 / p99 |
|---|---|---|---|---|---|---|---|---|---|---|
Fd7btgySsrju...Beauharnois, CA | 1 | 73 | 71 / 0 / 2 | -5.395 ms / -0.004 ms / 31.417 ms | 71 / 0 / 2 | -23 / -5 / 36 | 71 / 1 / 1 | -22 / -6 / 35 | 71 / 1 / 1 | -3 / -1 / 4 |
9eGrDohdNTAo...London, GB | 1 | 31 | 0 / 0 / 31 | 0.009 ms / 0.260 ms / 4.408 ms | 0 / 0 / 31 | 1 / 4 / 5 | 0 / 0 / 31 | 1 / 4 / 5 | 0 / 31 / 0 | 0 / 0 / 0 |
5pPRHniefFjk...Luxembourg, LU | 1 | 27 | 16 / 0 / 11 | -0.005 ms / 0.023 ms / 7.263 ms | 16 / 0 / 11 | -1 / 5 / 5 | 16 / 0 / 11 | -1 / 4 / 5 | 4 / 23 / 0 | 0 / 0 / 0 |
JD549HsbJHeE...London, GB | 1 | 25 | 0 / 0 / 25 | 0.009 ms / 0.052 ms / 0.112 ms | 0 / 0 / 25 | 1 / 3 / 7 | 0 / 0 / 25 | 1 / 3 / 7 | 0 / 25 / 0 | 0 / 0 / 0 |
CoG8d9Fp2TFJ...Singapore, SG | 1 | 11 | 11 / 0 / 0 | -0.009 ms / -0.005 ms / -0.005 ms | 11 / 0 / 0 | -7 / -2 / -2 | 11 / 0 / 0 | -8 / -3 / -3 | 10 / 1 / 0 | -1 / 0 / 0 |
ana2y2YvQ3ZP...Aubervilliers, FR | 1 | 11 | 3 / 0 / 8 | 0.011 ms / 0.574 ms / 0.574 ms | 3 / 0 / 8 | 2 / 35 / 35 | 3 / 0 / 8 | 2 / 6 / 6 | 0 / 9 / 2 | 0 / 1 / 1 |
7cVfgArCheMR...Amsterdam, NL | 1 | 7 | 4 / 0 / 3 | -0.009 ms / 0.012 ms / 0.012 ms | 4 / 0 / 3 | -1 / 7 / 7 | 2 / 3 / 2 | 0 / 5 / 5 | 0 / 6 / 1 | 0 / 1 / 1 |
HH5dA42XF1Hx...Amsterdam, NL | 1 | 2 | 2 / 0 / 0 | -0.005 ms / -0.005 ms / -0.005 ms | 2 / 0 / 0 | -1 / -1 / -1 | 2 / 0 / 0 | -1 / -1 / -1 | 0 / 2 / 0 | 0 / 0 / 0 |
mds4GEuiSgQR...Tokyo, JP | 1 | 2 | 2 / 0 / 0 | -20.209 ms / -20.209 ms / -20.209 ms | 2 / 0 / 0 | -54 / -54 / -54 | 2 / 0 / 0 | -48 / -48 / -48 | 2 / 0 / 0 | -5 / -5 / -5 |
2dfgsiSaZ51Q...Amsterdam, NL | 1 | 1 | 0 / 0 / 1 | 0.005 ms / 0.005 ms / 0.005 ms | 0 / 0 / 1 | 3 / 3 / 3 | 0 / 0 / 1 | 1 / 1 / 1 | 0 / 1 / 0 | 0 / 0 / 0 |
2mMGsb5uy1Q4...Stockholm, SE | 1 | 1 | 1 / 0 / 0 | -0.013 ms / -0.013 ms / -0.013 ms | 1 / 0 / 0 | -1 / -1 / -1 | 1 / 0 / 0 | -1 / -1 / -1 | 0 / 1 / 0 | 0 / 0 / 0 |
ACvL73V4GNnx...Luxembourg, LU | 1 | 1 | 1 / 0 / 0 | -0.004 ms / -0.004 ms / -0.004 ms | 1 / 0 / 0 | -1 / -1 / -1 | 1 / 0 / 0 | -1 / -1 / -1 | 0 / 1 / 0 | 0 / 0 / 0 |
EBk678aQvc3c...Los Angeles, US | 1 | 1 | 1 / 0 / 0 | -0.009 ms / -0.009 ms / -0.009 ms | 1 / 0 / 0 | -3 / -3 / -3 | 1 / 0 / 0 | -4 / -4 / -4 | 1 / 0 / 0 | -1 / -1 / -1 |
ECZx4Dfyn2o5...Bloomfield, US | 1 | 1 | 0 / 0 / 1 | 0.011 ms / 0.011 ms / 0.011 ms | 0 / 0 / 1 | 2 / 2 / 2 | 0 / 0 / 1 | 2 / 2 / 2 | 0 / 1 / 0 | 0 / 0 / 0 |
H8fHToVcZPi5...London, EN | 1 | 1 | 1 / 0 / 0 | -0.011 ms / -0.011 ms / -0.011 ms | 1 / 0 / 0 | -17 / -17 / -17 | 1 / 0 / 0 | -10 / -10 / -10 | 1 / 0 / 0 | -2 / -2 / -2 |
HW4zorvt6xDw...Hong Kong, HK | 1 | 1 | 0 / 0 / 1 | 0.005 ms / 0.005 ms / 0.005 ms | 0 / 0 / 1 | 1 / 1 / 1 | 0 / 0 / 1 | 1 / 1 / 1 | 0 / 1 / 0 | 0 / 0 / 0 |
RNXnAJV1DeBt...Prague, CZ | 1 | 1 | 0 / 0 / 1 | 0.011 ms / 0.011 ms / 0.011 ms | 0 / 0 / 1 | 1 / 1 / 1 | 0 / 0 / 1 | 1 / 1 / 1 | 0 / 1 / 0 | 0 / 0 / 0 |
SA1LFXr4os2P...São Paulo, BR | 1 | 1 | 1 / 0 / 0 | -26.560 ms / -26.560 ms / -26.560 ms | 1 / 0 / 0 | -40 / -40 / -40 | 1 / 0 / 0 | -45 / -45 / -45 | 1 / 0 / 0 | -5 / -5 / -5 |
SSmBEooM7Rkm...Amsterdam, NL | 1 | 1 | 0 / 0 / 1 | 0.013 ms / 0.013 ms / 0.013 ms | 0 / 0 / 1 | 1 / 1 / 1 | 0 / 0 / 1 | 1 / 1 / 1 | 0 / 1 / 0 | 0 / 0 / 0 |
kom1oNHyyt84...Amsterdam, NL | 1 | 1 | 0 / 0 / 1 | 0.011 ms / 0.011 ms / 0.011 ms | 0 / 0 / 1 | 1 / 1 / 1 | 0 / 0 / 1 | 1 / 1 / 1 | 0 / 1 / 0 | 0 / 0 / 0 |
mALL2W6DUgDD...São Paulo, BR | 1 | 1 | 1 / 0 / 0 | -17.509 ms / -17.509 ms / -17.509 ms | 1 / 0 / 0 | -111 / -111 / -111 | 1 / 0 / 0 | -77 / -77 / -77 | 1 / 0 / 0 | -5 / -5 / -5 |
Outcome triplets are candidate earlier / same / baseline earlier. Every percentile uses only complete same-slot values for that measure.
Methodology
Paired design
Both transactions share the leader run, slot trigger, sender host, transaction shape, and fee configuration. Only the network route differs.
Same-slot rule
First-observation time, processed transaction index, entry index, and PoH interval are compared only when both signatures land in the same slot.
Observation clock
Each signature uses its earliest event across the configured RPCEdge receivers. Deshred and processed receivers are not compared with each other.
Block placement
Processed transaction index is canonical transaction order. Entry index and preceding PoH ticks are derived after collection from the saved contiguous entry history.
Interpretation
Observation time includes sender scheduling, network transit, leader admission, block production, and propagation to the Frankfurt observer. It is not network RTT or TPU packet-arrival latency.
Geography
City, country, latitude, and longitude are historical values saved in getLeaderSlots snapshots. The coarse region classifier is not used.
The report does not claim that repeated transactions from one validator are independent samples or that geographic distance alone caused an observed difference.