- html - 出于某种原因,IE8 对我的 Sass 文件中继承的 html5 CSS 不友好?
- JMeter 在响应断言中使用 span 标签的问题
- html - 在 :hover and :active? 上具有不同效果的 CSS 动画
- html - 相对于居中的 html 内容固定的 CSS 重复背景?
我正在运行非常“简单”的测试。
@Fork(value = 1, jvmArgs = { "--illegal-access=permit", "-Xms10G", "-XX:+UnlockDiagnosticVMOptions", "-XX:+DebugNonSafepoints", "-XX:ActiveProcessorCount=7",
"-XX:+UseNUMA"
, "-XX:+UnlockDiagnosticVMOptions", "-XX:DisableIntrinsic=_currentTimeMillis,_nanoTime",
"-Xmx10G", "-XX:+UnlockExperimentalVMOptions", "-XX:ConcGCThreads=5", "-XX:ParallelGCThreads=10", "-XX:+UseZGC", "-XX:+UsePerfData", "-XX:MaxMetaspaceSize=10G", "-XX:MetaspaceSize=256M"}
)
@Benchmark
public String generateRandom() {
return UUID.randomUUID().toString();
}
可能不是很简单,因为使用随机,但同样的问题出现在使用 java 的任何其他测试上
Intel(R) Core(TM) i7-8700 CPU @ 3.20GHz 12 Threads (hyperthreading enabled ), 64 GB Ram, "Ubuntu" VERSION="20.04.2 LTS (Focal Fossa)"
Linux homepc 5.8.0-59-generic #66~20.04.1-Ubuntu SMP Thu Jun 17 11:14:10 UTC 2021 x86_64 x86_64 x86_64 GNU/Linux
7个线程的性能:
Benchmark Mode Cnt Score Error Units
RulesBenchmark.generateRandom thrpt 5 1312295.357 ± 27853.707 ops/s
Flame Graph with AsyncProfiler Result with 7 Thread At Home
Linux 5.4.17-2102.201.3.el8uek.x86_64 #2 SMP Fri Apr 23 09:05:57 PDT 2021 x86_64 x86_64 x86_64 GNU/Linux
Intel(R) Xeon(R) Gold 6258R CPU @ 2.70GHz with 56 Threads(hyperthreading disabled, the same when enabled and there is 112 cpu threads ) and 1 TB RAM I have half of performance (Even increasing threads) NAME="Oracle Linux Server" VERSION="8.4"
使用 1 个线程,我的性能非常出色:
Benchmark Mode Cnt Score Error Units
RulesBenchmark.generateRandom thrpt 5 2377471.113 ± 8049.532 ops/s
Flame Graph with AsyncProfiler Result 1 Thread
Benchmark Mode Cnt Score Error Units
RulesBenchmark.generateRandom thrpt 5 688612.296 ± 70895.058 ops/s
Flame Graph with AsyncProfiler Result 7 Thread
available: 1 nodes (0)
node 0 cpus: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55
node 0 size: 1030835 MB
node 0 free: 1011029 MB
node distances:
node 0
0: 10
但是在使用以下命令禁用一些 cpu 线程后:
for i in {12..55}
do
# your-unix-command-here
echo '0'| sudo tee /sys/devices/system/cpu/cpu$i/online
done
性能提升不大,不大。
.annobin___pthread_cond_signal.start
上
Oracle Linux
部署了 vagrant 镜像和我的家庭桌面上的内核版本并使用 10 个 cpu 线程运行它,性能几乎与我的桌面上相同(~1M op/sec)。所以这不是关于操作系统或内核,而是一些配置
最佳答案
本质上,您的基准测试了SecureRandom
的吞吐量。 .默认实现是synchronized (更准确地说,默认实现混合了输入表单 /dev/urandom
和上述提供程序)。
矛盾的是,更多的线程会导致更多的争用,从而降低整体性能,因为算法的主要部分无论如何都处于全局锁定之下。 Async-profiler 确实表明瓶颈是 Java 监视器上的同步:__lll_unlock_wake
, __pthread_cond_wait
, __pthread_cond_signal
- 都来自那个同步。
争用开销肯定取决于硬件、固件和操作系统配置。而不是试图减少这种开销(这可能很难,因为,你知道,有一天会出现另一个安全补丁,这将使系统调用慢 2 倍,例如),我建议首先摆脱争用地方。
这可以通过安装不同的非阻塞 SecureRandom
来实现。提供者如 this answer 中所示.我不会就特定的 SecureRandomSpi
给出建议,因为这取决于您的特定要求(吞吐量/可扩展性/安全性)。只会提到一个实现可以基于
/dev/urandom
访问等关于linux - Oracle Linux 上的 Java 性能问题,我们在Stack Overflow上找到一个类似的问题: https://stackoverflow.com/questions/68205982/
关闭。这个问题是off-topic .它目前不接受答案。 想要改进这个问题? Update the question所以它是on-topic用于堆栈溢出。 关闭 12 年前。 Improve thi
我有一个动态网格,其中的数据功能需要正常工作,这样我才能逐步复制网格中的数据。假设在第 5 行中,我输入 10,则从第 6 行开始的后续行应从 11 开始读取,依此类推。 如果我转到空白的第一行并输入
我有一个关于我的按钮消失的问题 我已经把一个图像作为我的按钮 用这个函数动画 function example_animate(px) { $('#cont
我有一个具有 Facebook 连接和经典用户名/密码登录的网站。目前,如果用户单击 facebook_connect 按钮,系统即可运行。但是,我想将现有帐户链接到 facebook,因为用户可以选
我有一个正在为 iOS 开发的应用程序,该应用程序执行以下操作 加载和设置注释并启动核心定位和缩放到位置。 map 上有很多注释,从数据加载不会花很长时间,但将它们实际渲染到 map 上需要一段时间。
我被推荐使用 Heroku for Ruby on Rails 托管,到目前为止,我认为我真的会喜欢它。只是想知道是否有人可以帮助我找出问题所在。 我按照那里的说明在该网站上创建应用程序,创建并提交
我看过很多关于 SSL 错误的帖子和信息,我自己也偶然发现了一个。 我正在尝试使用 GlobalSign CA BE 证书通过 Android WebView 访问网页,但出现了不可信错误。 对于大多
我想开始使用 OpenGL 3+ 和 4,但我在使用 Glew 时遇到了问题。我试图将 glew32.lib 包含在附加依赖项中,并且我已将库和 .dll 移动到主文件夹中,因此不应该有任何路径问题。
我已经盯着这两个下载页面的源代码看了一段时间,但我似乎找不到问题。 我有两个下载页面,一个 javascript 可以工作,一个没有。 工作:http://justupload.it/v/lfd7不是
我一直在使用 jQuery,只是尝试在单击链接时替换文本字段以及隐藏/显示内容项。它似乎在 IE 中工作得很好,但我似乎无法让它在 FF 中工作。 我的 jQuery: $(function() {
我正在尝试为 NDK 编译套接字库,但出现以下两个错误: error: 'close' was not declared in this scope 和 error: 'min' is not a m
我正在使用 Selenium 浏览器自动化框架测试网站。在测试过程中,我切换到特定的框架,我们将其称为“frame_1”。后来,我在 Select 类中使用了 deselectAll() 方法。不久之
我正在尝试通过 Python 创建到 Heroku PostgreSQL 数据库的连接。我将 Windows10 与 Python 3.6.8 和 PostgreSQL 9.6 一起使用。 我从“ht
我有一个包含 2 列的数据框,我想根据两列之间的比较创建第三列。 所以逻辑是:第 1 列 val = 3,第 2 列 val = 4,因此新列值什么都没有 第 1 列 val = 3,第 2 列 va
我想知道如何调试 iphone 5 中的 css 问题。 我尝试使用 firelite 插件。但是从纵向旋转到横向时,火石占据了整个屏幕。 有没有其他方法可以调试 iphone 5 中的 css 问题
所以我有点难以理解为什么这不起作用。我正在尝试替换我正在处理的示例站点上的类别复选框。我试图让它做以下事情:未选中时以一种方式出现,悬停时以另一种方式出现(选中或未选中)选中时以第三种方式出现(而不是
Javascript CSS 问题: 我正在使用一个文本框来写入一个 div。我使用以下 javascript 获取文本框来执行此操作: function process_input(){
你好,我很难理解 P、NP 和多项式时间缩减的主题。我试过在网上搜索它并问过我的一些 friend ,但我没有得到任何好的答案。 我想问一个关于这个话题的一般性问题: 设 A,B 为 P 中的语言(或
你好,我一直在研究 https://leetcode.com/problems/2-keys-keyboard/并想到了这个动态规划问题。 您从空白页上的“A”开始,完成后得到一个数字 n,页面上应该
我正在使用 Cocoapods 和 KIF 在 Xcode 服务器上运行持续集成。我已经成功地为一个项目设置了它来报告每次提交。我现在正在使用第二个项目并收到错误: Bot Issue: warnin
我是一名优秀的程序员,十分优秀!