zoukankan      html  css  js  c++  java
  • liblfds 测试

    liblfds 是一个灵活的无锁的数据结构库。支持 x86、x64、ARM 上的 Windows 和 Linux 系统。当前该数据结构库包含:列表、队列、ringbuffer、链接列表和堆栈。

    使用 liblfds 6.1.1 在一台32 核心的X86机器上测试了一下,这里贴出一部分,从数据看来,核心越多,每核心每秒操作数

    反而越低,总体性能曾下降趋势,怪了,顿时影响我对这个库的信心

    Benchmark Iteration 01
    ========================

    Release 6.1.1 Freelist Benchmark #1
    CPUs,total ops,mean ops/sec per CPU,standard deviation,scalability
    1,246081358,24608136,0,1.00
    2,199543112,9977156,11895333,0.41
    3,89551600,2985053,2899417,0.12
    4,72142158,1803554,1736669,0.07
    5,63174052,1263481,1231531,0.05
    6,56724522,945409,928488,0.04
    7,52423462,748907,741945,0.03
    8,49750856,621886,622482,0.03
    9,33216690,369074,885494,0.01
    10,29548254,295483,577371,0.01
    11,27488642,249897,349166,0.01
    12,27429622,228580,243030,0.01
    13,26364156,202801,198908,0.01
    14,26559260,189709,198769,0.01
    15,25467484,169783,200595,0.01
    16,25168734,157305,224077,0.01
    17,26750556,157356,224953,0.01
    18,26440556,146892,183104,0.01
    19,27767700,146146,182785,0.01
    20,28852032,144260,179046,0.01
    21,30215072,143881,177123,0.01
    22,31246522,142030,173322,0.01
    23,32409560,140911,168121,0.01
    24,33457132,139405,163690,0.01
    25,33074418,132298,197168,0.01
    26,32506334,125024,224112,0.01
    27,31940668,118299,256833,0.00
    28,31774470,113480,260138,0.00
    29,31559118,108825,259814,0.00
    30,31162320,103874,261511,0.00
    31,30791554,99328,265175,0.00
    32,30343526,94824,270157,0.00

    Release 6.1.1 Queue Benchmark #1
    CPUs,total ops,mean ops/sec per CPU,standard deviation,scalability
    1,77646406,7764641,0,1.00
    2,66672016,3333601,3548060,0.43
    3,50812638,1693755,1474220,0.22
    4,57219558,1430489,1308237,0.18
    5,57723612,1154472,1082222,0.15
    6,56615766,943596,895511,0.12
    7,53761898,768027,737010,0.10
    8,52140530,651757,632714,0.08
    9,23796974,264411,422512,0.03
    10,20895830,208958,409337,0.03
    11,20253188,184120,321570,0.02
    12,19833992,165283,227924,0.02
    13,19149290,147302,160678,0.02
    14,18602130,132872,133459,0.02
    15,17924544,119497,133588,0.02
    16,17623768,110149,147579,0.01
    17,17453130,102665,133979,0.01
    18,17135614,95198,112838,0.01
    19,17031684,89640,102983,0.01
    20,16904596,84523,93208,0.01
    21,16865270,80311,87486,0.01
    22,16764744,76203,82259,0.01
    23,16669194,72475,78501,0.01
    24,16607532,69198,75868,0.01
    25,16408878,65636,80149,0.01
    26,16213512,62360,84749,0.01
    27,16021986,59341,87562,0.01
    28,15915004,56839,86468,0.01
    29,15780098,54414,88629,0.01
    30,15660432,52201,90686,0.01
    31,15488272,49962,94192,0.01
    32,15325536,47892,97841,0.01

  • 相关阅读:
    第三篇:python函数
    第二篇:数据类型
    第一篇:初识python
    PyTorch教程之Autograd
    PyTorch教程之Tensors
    如何解决Python.h:No such file or directory
    如何解决conda install:command not found问题
    Linux 安装Anaconda 4.4.0
    Linux安装pytorch的具体过程以及其中出现问题的解决办法
    Writing Science 7.10 (The Opening and The Funnel)
  • 原文地址:https://www.cnblogs.com/jiayy/p/3014968.html
Copyright © 2011-2022 走看看