SIMULATION DRIVEN BY LAB MEASUREMENTS

See the difference yourself

Same task, two systems. Only the visuals are simulated; every duration and result was measured in our lab.

Data volume

Orders queue at peak times

Real time

BEFORE · Orders queue at peak times

Ready to replay

0 sec / 2.95 sec

0 / 256 records

AFTER · Orders queue at peak times

Ready to replay

0 sec / 0.381 sec

0 / 256 records

Recorded result256 records · Before: 2.95 sec · After: 0.381 sec · 7.7×

The screen and progress are illustrations; no query runs. Only recorded durations are replayed.

Duration by record count

Markers show recorded points; no values are inferred between them.

  • Before
  • After

Time (sec)

012316642562.90.4
012316642562.90.4

Inspect a recorded chart point

Record count · logarithmic axis

Show as a table
Duration by record count
Record countBeforeAfter
160.183 sec0.032 sec
640.73 sec0.097 sec
2562.95 sec0.381 sec

10 runs · median

Environment and method

Apple M3 · 8 cores · 24 GiB · macOS · native arm64 · shared developer workstation

16/64/256 synthetic orders requiring three independent service calls. Every complete result array matches an independent Python calculation.

Corretto 8u432 · Corretto 25.0.3 · -Xms128m -Xmx128m -XX:+UseG1GC · Python output oracle · 140 fresh JVM runs: 3 loads × 4 implementations × 10 repeats plus 20 constrained-service controls. At least 2s warmup, fixed randomized order.

Show measurement protocols

One order, three services. Stop stacking the waits.

Same Java 8 bytecode on both JDKs, identical synthetic outputs and dependency service. Sequential pool of 8, Java 25 runtime-only control and matched 64-call platform/virtual arms. Additional dependency-capacity-8 control; all 140 runs published. Primary point of 256 orders selected before measurement. Duration covers complete batch completion; platform threads are the median ThreadMXBean peak.

Measurement script ↗