===== Test Droplet with no load on CPU =====
==== Digital Ocean Sick $5/month droplet ====
john@mqtt1604testb:~$ ''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: 524.5674s
total number of events: 10000
total time taken by event execution: 524.0836
per-request statistics:
min: 6.48ms
avg: 52.41ms
max: 613.14ms
approx. 95 percentile: 125.51ms
Threads fairness:
events (avg/stddev): 10000.0000/0.00
execution time (avg/stddev): 524.0836/0.00
==== Digital Ocean Healthy $5/month droplet ====
john@mqtt1604C:~/mosBridge$ ''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: 12.8076s
total number of events: 10000
total time taken by event execution: 12.8054
per-request statistics:
min: 1.20ms
avg: 1.28ms
max: 2.52ms
approx. 95 percentile: 1.37ms
Threads fairness:
events (avg/stddev): 10000.0000/0.00
execution time (avg/stddev): 12.8054/0.00
Note: Jan 5, 2019 healthy $5/month droplet performance (verified on 6 droplets):
* ''execution time (avg/stddev): 15.3699/0.00''
==== Linode $5/month Droplet ====
sysbench 1.0.18 (using system LuaJIT 2.1.0-beta3)
Running the test with following options:
Number of threads: 1
Initializing random number generator from current time
Prime numbers limit: 10000
Initializing worker threads...
Threads started!
CPU speed:
events per second: 1672.13
General statistics:
total time: 10.0001s
total number of events: 16723
Latency (ms):
min: 0.58
avg: 0.60
max: 3.55
95th percentile: 0.62
sum: 9994.08
Threads fairness:
events (avg/stddev): 16723.0000/0.00
execution time (avg/stddev): 9.9941/0.00
==== AWS $5/month Droplet ====
ubuntu@ip-172-26-9-190:~$ sysbench --test=cpu run
WARNING: the --test option is deprecated. You can pass a script name or path on the command line without any options.
sysbench 1.0.18 (using system LuaJIT 2.1.0-beta3)
Running the test with following options:
Number of threads: 1
Initializing random number generator from current time
Prime numbers limit: 10000
Initializing worker threads...
Threads started!
CPU speed:
events per second: 767.07
General statistics:
total time: 10.0008s
total number of events: 7673
Latency (ms):
min: 1.22
avg: 1.30
max: 1.63
95th percentile: 1.32
sum: 9986.47
Threads fairness:
events (avg/stddev): 7673.0000/0.00
execution time (avg/stddev): 9.9865/0.00
==== Measure CPU load on all Droplets: ====
''./sshToServer.sh john all "top -bn1 | grep 'Cpu(s)' | sed 's/.*, *\([0-9.]*\)%* id.*/\1/' | awk '{print 100 - \$1}'"''
it seems more accurate (or at least lines up better with DO usage stats) if a bit slower to catch the 2nd iteration of top instead of the first:
''./sshToServer.sh john all "top -bn2 | grep 'Cpu(s)' | sed 's/.*, *\([0-9.]*\)%* id.*/\1/' | awk '{print 100 - \$1}' | tail -n 1"''