예전에 2651v2 48 코어에서 돌린거 시간을 저장을 안해놔서 아쉽네..
# sysbench --test=cpu --cpu-max-prime=200000 --num-threads=96 run sysbench 0.4.12: multi-threaded system evaluation benchmark Running the test with following options: Number of threads: 96 Doing CPU performance benchmark Threads started! Done. Maximum prime number checked in CPU test: 200000 Test execution summary: total time: 6.8619s total number of events: 10000 total time taken by event execution: 652.8527 per-request statistics: min: 65.21ms avg: 65.29ms max: 129.36ms approx. 95 percentile: 65.30ms Threads fairness: events (avg/stddev): 104.1667/0.42 execution time (avg/stddev): 6.8005/0.03 |
# sysbench --test=cpu --cpu-max-prime=200000 --num-threads=48 run sysbench 0.4.12: multi-threaded system evaluation benchmark Running the test with following options: Number of threads: 48 Doing CPU performance benchmark Threads started! Done. Maximum prime number checked in CPU test: 200000 Test execution summary: total time: 13.6361s total number of events: 10000 total time taken by event execution: 652.2536 per-request statistics: min: 65.20ms avg: 65.23ms max: 66.83ms approx. 95 percentile: 65.26ms Threads fairness: events (avg/stddev): 208.3333/0.47 execution time (avg/stddev): 13.5886/0.03 |
'embeded > ARM' 카테고리의 다른 글
ampere altra (0) | 2020.11.23 |
---|---|
arm cl (0) | 2020.10.29 |
cavium thunderX / thunderX2 (0) | 2020.09.18 |
thunderX 아키텍쳐 (0) | 2020.09.16 |
keil window 버전별 지원버전 (0) | 2018.07.19 |