随笔分类 - Performance
摘要:问题1: 15ms 内多次调用 GetSystemTime / GetLocalTime(java 中对应的函数为 System.currentTimeMillis()), 返回相同的值 解决办法, 使用GetSystemTime 作为baseline, 然后用windows 提供的高精度计时器 QueryPerformanceCounter(java 中相应的函数为 System.nanoTime()) 做计时,精确时钟为baseline + 计时器时间。问题2: QueryPerformanceCounter/QueryPerformanceFrequency 的问题这个问题主要取决与wi
阅读全文
摘要:Windows API:GetSystemTime/GetLocalTimeJAVA:Calendar cal = new GregorianCalendar(TimeZone.getTimeZone("GMT"));15 毫秒内调用多次, 只能得到相同的timestamp
阅读全文
摘要:http://engin.bzzzt.biz/embeddeddb.shtmlUPDATE 20-OCT-2008:Instead of creating a DBCommand for each iteration, I started to use one DBCommand instance, as any normal person would do in the first place. And it chagned the results quite a bit. It looks like object initiation for Firebird is expensive..
阅读全文
摘要:http://www.cs.rochester.edu/research/synchronization/
阅读全文
摘要:http://www.cs.rochester.edu/research/synchronization/pseudocode/queues.html/Files/slime/1996_PODC_queues.pdfPseudocode from article of the above name in PODC96 (with two typos corrected), by Maged M. ...
阅读全文
摘要:The answer depends on what exactly you're trying to accelerate.GPU is really a "software acceleration" that can accelerate certain class of compute-intensive applications.The only interface to the G...
阅读全文
摘要:operating system 64 bit systems have several advantages over 32 bit systems: The number of registers is doubled. This makes it possible to store intermediate data and local variables in registers rath...
阅读全文
摘要:http://www.agner.org/optimize/非常好的程序优化相关的resource
阅读全文
摘要:There is a significant runtime cost for position-independent code when using dynamic linking (shared objects) in Linux, BSD and Mac systems. Critical CPU-intensive code should never be stored in a sha...
阅读全文
摘要:http://hi.baidu.com/spkiller/blog/item/b216a80176137f00728da57f.html完成端口理论值socket上限是6万多,也就是说纯连接能达到65535个,而select只能64个,更改了FD_MAXSIZE后最多1024,但是实际只能支撑七八百的极限。 由于某天对Winsock select模型的一个奇怪猜想,google、baidu了半天...
阅读全文
摘要:Dr. Insung Park and Alex BendetovWelcome back for the second part of our two-part article series: Core Instrumentation Events in Windows 7. In the first article, we presented a high-level overview ...
阅读全文
摘要:Dr. Insung Park and Alex BendetoversToday's computer software constantly breaks new grounds. Consumer software applications offer a sophisticated set of features that enable rich new experiences. P...
阅读全文
摘要:http://blogs.msdn.com/b/ryanmy/archive/2005/05/27/422772.aspx Uniformity. If you're debugging systemic problems involving multiple components, and all the involved components use ETW, you can have the...
阅读全文
摘要:1 create and write the manifest.xml代码Code highlighting produced by Actipro CodeHighlighter (freeware)http://www.CodeHighlighter.com/--<instrumentationManifestxmlns="http://schemas.microsoft.com/wi...
阅读全文
摘要:代码Code highlighting produced by Actipro CodeHighlighter (freeware)http://www.CodeHighlighter.com/--//EventLogging.cpp:Definestheentrypointfortheconsoleapplication.//#include"stdafx.h"#include<wind...
阅读全文
摘要:http://blogs.technet.com/b/winserverperformance/archive/2008/06/26/designing-applications-for-high-performance-part-iii.aspx25 Jun 2008 7:44 PM Comments 14 Rick Vicik - Architect, Windows Server Perfo...
阅读全文
摘要:http://blogs.technet.com/b/winserverperformance/archive/2008/05/21/designing-applications-for-high-performance-part-ii.aspx 20 May 2008 9:48 PM Comments 1 Rick Vicik - Architect, Windows Server Perfo...
阅读全文
摘要:http://blogs.technet.com/b/winserverperformance/archive/2008/04/25/designing-applications-for-high-performance-part-1.aspx25 Apr 2008 12:40 AM Comments 1 Rick Vicik - Architect, Windows Server Perfor...
阅读全文
摘要:msdn states that InterlockedCompareExchange has performance penalties due to "memory fences". For performance gains modern CPUs often execute instructions out of order to make maximum use of th...
阅读全文
摘要:http://msdn.microsoft.com/zh-cn/magazine/cc163437.aspx事件跟踪 使用 ETW 改善调试和性能优化 Dr. Insung Park andRicky Buch 本文讨论: 了解 ETW 体系结构Windows Vista 中 ETW 的新功能使用事件提供程序 API 进行编程 本文使用了以下技术: Windows Vista 目录 Wind...
阅读全文

浙公网安备 33010602011771号