The PHP Bench
All runs

20260928011232-b9eaed5b

complete Signed by kaiju · 9f9bb07dedd6...91a4d3d3
Started
Generated
Measured samples
1,872
Combinations
13

Machine

Captured from the runner's system manifest.

Host kaiju Linux · kernel 6.19.8-200.nobara.fc43.x86_64
PHP 8.5.11
Processor Intel(R) Xeon(R) CPU E5-2687W v4 @ 3.00GHz 48 cores · x86_64
Memory 251.7 GB P1_DIMMA1 32 GB 2133 MT/s DDR4 M386A4G40DM0-CPB P1_DIMMB1 32 GB 2133 MT/s DDR4 M386A4G40DM0-CPB P1_DIMMC1 32 GB 2133 MT/s DDR4 M386A4G40DM0-CPB P1_DIMMD1 32 GB 2133 MT/s DDR4 M386A4G40DM0-CPB P2_DIMME1 32 GB 2133 MT/s DDR4 M386A4G40DM0-CPB P2_DIMMF1 32 GB 2133 MT/s DDR4 M386A4G40DM0-CPB P2_DIMMG1 32 GB 2133 MT/s DDR4 M386A4G40DM0-CPB P2_DIMMH1 32 GB 2133 MT/s DDR4 M386A4G40DM0-CPB
Disk 447.1 GB Device: /dev/sda Capacity: 447.1 GB

Runtime envelope

End-to-end time per scenario run across all non-warmup samples on mariadb.

Min sample
85.0 ms
Mean sample
1,198 ms
Max sample
8,144 ms
Cumulative time
37m 22s
Throughput
0.835/s

Runtime comparison

Mean time per scenario run in this upload, warmup runs excluded, sorted by runtime.

Init Insert Read Update Delete
  1. PHPFUIBatch 3.0.3 mariadb 627 ms
  2. Doctrine 3.6.2 mariadb 829 ms
  3. Cycle 2.16.0 mariadb 936 ms
  4. Atlas 2.0.1 mariadb 941 ms
  5. PHPFUI 3.0.3 mariadb 1,050 ms
  6. Propel2 6cb272b6 mariadb 1,091 ms
  7. CakeCached 5.3.3 mariadb 1,168 ms
  8. DivergenceV2 2.1.4 mariadb 1,170 ms
  9. Cake 5.3.3 mariadb 1,187 ms
  10. DivergenceV3 3.2.0 mariadb 1,307 ms
  11. Yii 2.0.54 mariadb 1,336 ms
  12. RedBean 5.7.5 mariadb 1,686 ms
  13. Eloquent 13.2.0 mariadb 2,242 ms

Where the time goes

Each bar is one framework's mean scenario run on mariadb, split into its phases and sorted by total. Bars share a scale within each engine. Hover a bar for the numbers.

Init Insert Read Update Delete
Mean split across 13 combinations
Insert 20% Read 10% Update 44% Delete 21%

mariadb

  1. PHPFUIBatch3.0.3 627 ms
  2. Doctrine3.6.2 829 ms
  3. Cycle2.16.0 936 ms
  4. Atlas2.0.1 941 ms
  5. PHPFUI3.0.3 1,050 ms
  6. Propel26cb272b6 1,091 ms
  7. CakeCached5.3.3 1,168 ms
  8. DivergenceV22.1.4 1,170 ms
  9. Cake5.3.3 1,187 ms
  10. DivergenceV33.2.0 1,307 ms
  11. Yii2.0.54 1,336 ms
  12. RedBean5.7.5 1,686 ms
  13. Eloquent13.2.0 2,242 ms
Exact numbers
PHPFUIBatch 3.0.3 mariadb 144 627 ms 14.4 ms 95.1 ms 463 ms 52.5 ms
Doctrine 3.6.2 mariadb 144 829 ms 67.7 ms 147 ms 386 ms 209 ms
Cycle 2.16.0 mariadb 144 936 ms 228 ms 70.2 ms 374 ms 257 ms
Atlas 2.0.1 mariadb 144 941 ms 232 ms 77.4 ms 373 ms 257 ms
PHPFUI 3.0.3 mariadb 144 1,050 ms 238 ms 92.4 ms 451 ms 266 ms
Propel2 6cb272b6 mariadb 144 1,091 ms 282 ms 15.2 ms 428 ms 361 ms
CakeCached 5.3.3 mariadb 144 1,168 ms 263 ms 198 ms 568 ms 134 ms
DivergenceV2 2.1.4 mariadb 144 1,170 ms 269 ms 111 ms 520 ms 267 ms
Cake 5.3.3 mariadb 144 1,187 ms 271 ms 202 ms 575 ms 133 ms
DivergenceV3 3.2.0 mariadb 144 1,307 ms 280 ms 154 ms 598 ms 273 ms
Yii 2.0.54 mariadb 144 1,336 ms 268 ms 135 ms 563 ms 366 ms
RedBean 5.7.5 mariadb 144 1,686 ms 471 ms 106 ms 772 ms 329 ms
Eloquent 13.2.0 mariadb 144 2,242 ms 363 ms 273 ms 1,037 ms 552 ms

Scenario families

Mean across every framework and engine in this upload.

  • simple 178 ms
  • string_variable 1,076 ms
  • string_fixed 1,108 ms
  • float_fixed 1,283 ms
  • int_fixed 1,296 ms
  • canary 1,546 ms

What the workloads measure

The row shape and operation mix decide where each framework spends its time.

Scenario families
simple baseline row shape
Shows framework overhead with minimal field complexity, so object creation and query setup dominate.
canary mixed field and nullable stress
Wide mixed records amplify casting, hydration, dirty tracking, and adapter metadata overhead.
int_fixed pure scalar writes
A low-noise floor. Slow results here usually mean framework machinery is expensive even before text and cast-heavy payloads show up.
float_fixed numeric conversion paths
Decimal and float normalization can widen update and verification cost, especially for ORMs that compare serialized values.
string_fixed predictable text hydration
Column count and text assignment matter more than payload variability, which makes hydration strategy easier to isolate.
string_variable application-like text workload
Variable strings and note fields make this the closest to real CRUD traffic and usually punish heavier active-record layers the most.
Operation shapes
Warmup vs hot runs
Run 1 is warmup. Hot-path averages exclude it so boot and metadata setup do not dominate steady-state rankings.
Insert and indexed schemas
Indexed tables make writes cost more because every insert and many updates must maintain secondary indexes as well as the row itself.
Read all vs read probes
Read All rewards bulk hydration efficiency; random and tail probes expose per-query setup, lookup overhead, and tiny-result inefficiency.
Update with verification
Each update shape is followed by a verification read, so dirty-checking, write SQL generation, hydration, and comparison all affect the final time.
Delete with confirmation
Deletes are checked afterward. Frameworks that pay extra for follow-up lookups or teardown lose more time here than a bare SQL path.
Total runtime
The total runtime metric is effectively a benchmark of the machine plus adapter behavior across the full workload, not just one CRUD verb.