The tests are the product. Watch them run.
One shared run measures every substrate primitive on this box, over a bounded slice of live public data, and streams each result here the instant it lands. Everyone watching sees the same run, so a full gallery of spectators adds nothing to the load. Nothing on this page is typed in by hand, if a tile reads measuring… the number has not landed yet.
graph traversal · the record + its neighbors, over the live value-flow
468.3MTEPS
warm working-set BFS over the resident value-flow graph, the trailing average across live runs
The same figure and the same method label the scorecard, the reproducibility API and the
graph demo all now read from, so the number never disagrees with itself.
see the proof ▸
Writes
Cold
cold
42.2M/s
cold durable append · O_DIRECT · single-shard
measured
94.7 ms
replaces a bolt-on write-ahead log
Reads
Cold
cold
62.7M/s
by-id coalesced · QD64 · O_DIRECT · single-shard
measured
25.5 ms
replaces a page cache you must pre-warm
Warm
warm
188.6M/s
random by-id gather · resident · single-shard
measured
17.0 ms
replaces an in-memory cache tier
Hot
hot
230.0M/s
random by-id gather · L3-resident · single-shard
measured
3.5 ms
replaces an in-memory cache tier
overall
blended
62.7M/s
effective rate at the measured tier mix
cold 100.0% · warm 0.0% · hot 0.0%
Traversals
Cold
cold
195.8MTEPS
cold first-touch · dense seeds · all-cores
measured
38.35 s
replaces a bolt-on native-graph store
Warm
warm
473.9MTEPS
page-cache-warm rerun · dense seeds · all-cores
measured
30.1 ms
replaces a bolt-on native-graph store
Hot
hot
2.19BTEPS
L3-local co-located image · dense seeds · all-cores
measured
1.3 ms
replaces a bolt-on native-graph store
overall
blended
195.8MTEPS
effective rate at the measured tier mix
cold 100.0% · warm 0.0% · hot 0.0%
Scans
Cold
cold
11.68GB/s
sequential sweep · QD64 · O_DIRECT · single-shard
measured
8.8 ms
replaces a columnar analytics engine
Warm
warm
36.43GB/s
resident SIMD sweep · in-process · single-shard
measured
3.5 ms
replaces a columnar analytics engine
Hot
hot
53.46GB/s
L3-resident SIMD sweep · in-process · single-shard
measured
1.9 ms
replaces a columnar analytics engine
overall
blended
12GB/s
effective rate at the measured tier mix
cold 100.0% · warm 0.0% · hot 0.0%
Binary-neuron kernel
Kernel
compute
3.5TSAOPS/s
all-core aggregate · 7 cores · AVX-512 VPOPCNTQ
measured
replaces a GPU inference tier
Conveyor · predictive prefetch · janitor efficiency, measured on the live handle
Conveyor · where the reads land
0.0%served at cache speed (warm + hot)
cold 100.0%
warm 0.0%
hot 0.0%
the cold band shrinks toward warm and hot as the conveyor lifts the working set, cache-speed reads over a disk-capacity dataset
Predictive prefetch · bounded band
100.0Krecords streamed into the walk
only the slice the query touches is prefetched, the pipe stays small while the reads stay hot, never the whole graph
Janitor · co-location, epoch over epoch
co-location warming · no epochs on this handle during the re-ingest yet
Every number here is measured live on this box, on a bounded slice of public Bitcoin data you
can re-pull over Bitcoin P2P and re-run. Writes go to a throwaway scratch area, never a live
store; reads, traversals and scans read the live value-flow graph without touching it. The
honesty band on each tile tells you what class of number it is.