一个ASP.NET性能瓶颈之谜 [英] an ASP.NET performance bottleneck mystery

问题描述

一个典型的ASP.NET应用程序 - APPSRV + MS SQL数据库。这两个服务器都采用重量举重8个内核，20 GB的RAM。当负载测试，吞吐量去某处400 VirtualUsers（根据LoadRunner的），以CPU为约30％利用的数据库服务器主要空转 - 响应时间急剧上升，给反应迟钝点

A classic ASP.NET app - AppSrv + MS SQL DB. Both servers are heavy-lifters 8 cores, 20 GB of RAM. When load testing, the throughput goes somewhere to 400 VirtualUsers (according to LoadRunner) with CPU being approximately 30% utilized an DB server primarily idling - response times go dramatically up, to the point of unresponsive.

通常嫌疑犯，如最大池耗尽和ASP.NET集康恩限制是没有过错：最大池设置为200，用于约80是conns;康恩极限设置为0

The usual suspects, such as Max Pool being exhausted and conn limit on ASP.NET set are not at fault: Max Pool is set to 200 and about 80 conns are used; conn limit is set to 0.

我跑了蚂蚁探查的code，它表明线程阻塞没有显著贡献。

I ran with ANTS profiler the code and it showed that Thread blocking did not contribute significantly.

想法非常非常欢迎！

推荐答案

是否主机使用的应用程序池？

Does the host utilize App Pool?

你有没有尝试在增加数5〜10

Did you try increase the number to 5 to 10 in

An Application Pool -> Performance -> 
Web Garden -> Max Number of worker processes

这篇关于一个ASP.NET性能瓶颈之谜的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！