Raspberry Pi 2 overclock benchmark

gistfile1.txt

# Procedure
#
# From a completely fresh restart, ssh'd into the Pi from my mac and then ran the 
# sysbench command listed at the top of each section. Copied and pasted the
# output, restarted, and ran the next.
#
# 1000MHz overclock settings:
#
# arm_freq=1000
# sdram_freq=500
# core_freq=500
# over_voltage=2
#
# 1100MHz overclock settings:
#
# arm_freq=1100
# sdram_freq=500
# core_freq=500
# over_voltage=6
#
# It looks like things may have got a little unstable at 1100MHz as the times
# were actually *higher* for all four cores. You can see the max request in the
# last test was almost 26 seconds.

################################################################################
# No overclock
################################################################################

pi@raspberrypi ~ $ sysbench --test=cpu run
sysbench 0.4.12:  multi-threaded system evaluation benchmark

Running the test with following options:
Number of threads: 1

Doing CPU performance benchmark

Threads started!
Done.

Maximum prime number checked in CPU test: 10000


Test execution summary:
    total time:                          299.0055s
    total number of events:              10000
    total time taken by event execution: 298.9891
    per-request statistics:
         min:                                 29.81ms
         avg:                                 29.90ms
         max:                                 51.81ms
         approx.  95 percentile:              29.97ms

Threads fairness:
    events (avg/stddev):           10000.0000/0.00
    execution time (avg/stddev):   298.9891/0.00

--------------------------------------------------------------------------------

pi@raspberrypi ~ $ sysbench --test=cpu --num-threads=4 run
sysbench 0.4.12:  multi-threaded system evaluation benchmark

Running the test with following options:
Number of threads: 4

Doing CPU performance benchmark

Threads started!
Done.

Maximum prime number checked in CPU test: 10000


Test execution summary:
    total time:                          74.5663s
    total number of events:              10000
    total time taken by event execution: 298.1804
    per-request statistics:
         min:                                 29.64ms
         avg:                                 29.82ms
         max:                                 45.07ms
         approx.  95 percentile:              29.96ms

Threads fairness:
    events (avg/stddev):           2500.0000/10.65
    execution time (avg/stddev):   74.5451/0.01

################################################################################
# 1000MHz overclock
################################################################################

pi@raspberrypi ~ $ sysbench --test=cpu run
sysbench 0.4.12:  multi-threaded system evaluation benchmark

Running the test with following options:
Number of threads: 1

Doing CPU performance benchmark

Threads started!
Done.

Maximum prime number checked in CPU test: 10000


Test execution summary:
    total time:                          267.0090s
    total number of events:              10000
    total time taken by event execution: 266.9946
    per-request statistics:
         min:                                 26.68ms
         avg:                                 26.70ms
         max:                                 50.54ms
         approx.  95 percentile:              26.73ms

Threads fairness:
    events (avg/stddev):           10000.0000/0.00
    execution time (avg/stddev):   266.9946/0.00

--------------------------------------------------------------------------------

pi@raspberrypi ~ $ sysbench --test=cpu --num-threads=4 run
sysbench 0.4.12:  multi-threaded system evaluation benchmark

Running the test with following options:
Number of threads: 4

Doing CPU performance benchmark

Threads started!
Done.

Maximum prime number checked in CPU test: 10000


Test execution summary:
    total time:                          67.0714s
    total number of events:              10000
    total time taken by event execution: 268.2266
    per-request statistics:
         min:                                 26.68ms
         avg:                                 26.82ms
         max:                                 45.09ms
         approx.  95 percentile:              26.93ms

Threads fairness:
    events (avg/stddev):           2500.0000/8.69
    execution time (avg/stddev):   67.0566/0.01

################################################################################
# 1100MHz overclock
################################################################################

pi@raspberrypi ~ $ sysbench --test=cpu run
sysbench 0.4.12:  multi-threaded system evaluation benchmark

Running the test with following options:
Number of threads: 1

Doing CPU performance benchmark

Threads started!
Done.

Maximum prime number checked in CPU test: 10000


Test execution summary:
    total time:                          244.0507s
    total number of events:              10000
    total time taken by event execution: 244.0377
    per-request statistics:
         min:                                 24.38ms
         avg:                                 24.40ms
         max:                                 45.50ms
         approx.  95 percentile:              24.45ms

Threads fairness:
    events (avg/stddev):           10000.0000/0.00
    execution time (avg/stddev):   244.0377/0.00

--------------------------------------------------------------------------------

pi@raspberrypi ~ $ sysbench --test=cpu --num-threads=4 run
sysbench 0.4.12:  multi-threaded system evaluation benchmark

Running the test with following options:
Number of threads: 4

Doing CPU performance benchmark

Threads started!
Done.

Maximum prime number checked in CPU test: 10000


Test execution summary:
    total time:                          86.8193s
    total number of events:              10000
    total time taken by event execution: 347.1923
    per-request statistics:
         min:                                 24.25ms
         avg:                                 34.72ms
         max:                              25866.72ms
         approx.  95 percentile:              24.45ms

Threads fairness:
    events (avg/stddev):           2500.0000/8.28
    execution time (avg/stddev):   86.7981/0.01