http://blogs.msdn.com/vancem/archive/2007/04/09/a-model-for-cold-startup-time-of-an-application-on-windows.aspxWell it is has been a while. I see now that it is been getting close to 1/2 a year since I last posted. Sigh. It is so easy for other things to get busy and not to blog. I have resolved to try to be a bit more methodical about and insure that I writeen-USCommunityServer 2.1 SP1 (Build: 61025.2)http://blogs.msdn.com/vancem/archive/2007/04/09/a-model-for-cold-startup-time-of-an-application-on-windows.aspx#2059958Mon, 09 Apr 2007 14:22:56 GMT91d46819-8472-40ad-a661-2c78acb4018c:2059958AtulGupta<p>Interesting in-sight. Looking forward to more information</p> http://blogs.msdn.com/vancem/archive/2007/04/09/a-model-for-cold-startup-time-of-an-application-on-windows.aspx#2091362Thu, 12 Apr 2007 00:03:42 GMT91d46819-8472-40ad-a661-2c78acb4018c:2091362dimkaz<p>Hi Vance,</p> <p>are you sure the following is not a big oversimplification?</p> <p>ColdStartup = WarmStartup + DiskIOTime.</p> <p>Considering that most io done on starup (such as dll loading) is done synch. and given multiple threads.</p> <p>As part of optimizing our startup we have the following split. (simplified) </p> <p>We break our startup into multiple thread that touch different code (dlls).</p> <p>On warm startup our 2x cores on the startup are almost at 100%.</p> <p>On the cold startup cpu utilization goes rarely to 60%. I should also note that we are not using ngen and the jit time is a factor here as well (increase CPU time)</p> <p>My rough estimate is that in my case you would be off by as much as 20%.</p> <p>What other scenarios do you consider? Is ngen case your main focus? The good old technique of ordering functions is gone from .net, will it return? What about metadata? </p> <p>Please continue to post. I always enjoy your articles.</p> http://blogs.msdn.com/vancem/archive/2007/04/09/a-model-for-cold-startup-time-of-an-application-on-windows.aspx#2147526Mon, 16 Apr 2007 01:22:49 GMT91d46819-8472-40ad-a661-2c78acb4018c:2147526Vance Morrison's Weblog<p>After my last blog entry on cold startup a reader ( dimkaz ) worried that the formula would not be accurate</p> http://blogs.msdn.com/vancem/archive/2007/04/09/a-model-for-cold-startup-time-of-an-application-on-windows.aspx#8497944Tue, 13 May 2008 03:59:29 GMT91d46819-8472-40ad-a661-2c78acb4018c:8497944Vance Morrison's Weblog<p>What's Coming in .NET Runtime Performance in Version V3.5 SP1 It certainly has been a while since I last</p>