I’ve the usual mac mini m2 professional (10-core cpu, 512gb ssd, 16GB ram). I even have a mac studio m1 max (no longer extremely!) 32gb ram, upped to 1TB ssd. each run macos 13.2.1. I then brew set up sysbench and
sysbench fileio --file-total-size=15G --file-test-mode=rndrw --time=300 --max-requests=0 run
sysbench fileio --file-total-size=15G --file-test-mode=rndrw --time=300 --max-requests=0 cleanup
sysbench 1.0.20 (the use of device LuaJIT 2.1.0-beta3)
the mini does
Operating the check with following choices:
Collection of threads: 1
Initializing random quantity generator from present time
Additional dossier open flags: (none)
128 information, 120MiB each and every
15GiB whole dossier length
Block length 16KiB
Collection of IO requests: 0
Learn/Write ratio for mixed random IO check: 1.50
Periodic FSYNC enabled, calling fsync() each and every 100 requests.
Calling fsync() on the finish of check, Enabled.
The usage of synchronous I/O mode
Doing random r/w check
Initializing employee threads...
Threads began!
Record operations:
reads/s: 8668.41
writes/s: 5778.94
fsyncs/s: 18492.97
Throughput:
learn, MiB/s: 135.44
written, MiB/s: 90.30
Normal statistics:
whole time: 300.0003s
whole selection of occasions: 9881989
Latency (ms):
min: 0.00
avg: 0.03
max: 20.48
ninety fifth percentile: 0.00
sum: 299269.99
Threads equity:
occasions (avg/stddev): 9881989.0000/0.00
execution time (avg/stddev): 299.2700/0.00
sysbench 1.0.20 (the use of device LuaJIT 2.1.0-beta3)
the studio does
Operating the check with following choices:
Collection of threads: 1
Initializing random quantity generator from present time
Additional dossier open flags: (none)
128 information, 120MiB each and every
15GiB whole dossier length
Block length 16KiB
Collection of IO requests: 0
Learn/Write ratio for mixed random IO check: 1.50
Periodic FSYNC enabled, calling fsync() each and every 100 requests.
Calling fsync() on the finish of check, Enabled.
The usage of synchronous I/O mode
Doing random r/w check
Initializing employee threads...
Threads began!
Record operations:
reads/s: 62863.67
writes/s: 41909.11
fsyncs/s: 134109.31
Throughput:
learn, MiB/s: 982.24
written, MiB/s: 654.83
Normal statistics:
whole time: 300.0010s
whole selection of occasions: 71664796
Latency (ms):
min: 0.00
avg: 0.00
max: 8.05
ninety fifth percentile: 0.00
sum: 294415.57
Threads equity:
occasions (avg/stddev): 71664796.0000/0.00
execution time (avg/stddev): 294.4156/0.00
sysbench 1.0.20 (the use of device LuaJIT 2.1.0-beta3)
Do I’ve one thing misconfigured at the m2 mini right here? The studio professional has learn throughput about 6-8 occasions sooner than the mini. I’ve heard that the bottom mini m2 used to be slower, however this gradual??