随笔分类 -  Performance

compiler optimization
摘要：https://en.wikipedia.org/wiki/Optimizing_compiler 阅读全文
posted @ 2023-05-12 19:03 fndefbwefsowpvqfx
openjdk Performance
摘要：https://wiki.openjdk.org/display/HotSpot/PerformanceTechniques Performance techniques used in the Hotspot JVM What code shapes does the JVM optimize b 阅读全文
posted @ 2023-05-12 19:01 fndefbwefsowpvqfx
atomic Herb Sutter
摘要：![](https://img2022.cnblogs.com/blog/859364/202204/859364-20220426184152960-376026084.png) ![](https://img2022.cnblogs.com/blog/859364/202204/859364-20220426184200877-736472788.png) ![](https://img202 阅读全文
posted @ 2022-04-26 18:44 fndefbwefsowpvqfx
vtable
摘要：![](https://img2022.cnblogs.com/blog/859364/202204/859364-20220426161627959-1346273786.png) ![](https://img2022.cnblogs.com/blog/859364/202204/859364-20220426161643252-272540731.png) ![](https://img20 阅读全文
posted @ 2022-04-26 16:18 fndefbwefsowpvqfx
CPU store buffer load buffer
摘要：![](https://img2022.cnblogs.com/blog/859364/202204/859364-20220426121803829-1653385920.png) ![](https://img2022.cnblogs.com/blog/859364/202204/859364-20220426122033459-630590719.png) ![](https://img20 阅读全文
posted @ 2022-04-26 12:22 fndefbwefsowpvqfx
variant overhead
摘要：![](https://img2022.cnblogs.com/blog/859364/202204/859364-20220426102239782-1160817732.png) ![](https://img2022.cnblogs.com/blog/859364/202204/859364-20220426104818392-237069994.png) ![](https://img20 阅读全文
posted @ 2022-04-26 10:23 fndefbwefsowpvqfx
cpu architecture
摘要：![](https://img2022.cnblogs.com/blog/859364/202204/859364-20220425144725524-992255343.png) 阅读全文
posted @ 2022-04-25 14:48 fndefbwefsowpvqfx
asynchronous disk I/O
摘要：Friday, October 26th, 2012 by arvid Since 2010, I’ve been working, on and off, on a branch off of libtorrent which use asynchronous disk I/O, instead 阅读全文
posted @ 2022-04-21 18:32 fndefbwefsowpvqfx
Performance
摘要：Memory alignment efficient memory load locality time space data dependency vectorize cache line 多线程共享数据，避免cache line 重叠, false sharing stack prefaulti 阅读全文
posted @ 2022-04-20 12:12 fndefbwefsowpvqfx
web page performance
摘要：整个web页面优化，涉及了传输，运行，渲染中的每个步骤。 自然每一步都有优化的可能性。 本质上使用的方法还是很朴素： 缓存(时、空局部性) 批量并发 预测(提前执行可能要执行的操作) 类似方法在CPU的指令运行(cache,pipeline)，网络协议栈(累积发送)中都大量存在。 甚至公司、团体的组 阅读全文
posted @ 2020-11-06 14:44 fndefbwefsowpvqfx