Flutterby™! : Microsoft TCP/IP benchmark speed explained

Next unread comment / Catchup all unread comments User Account Info | Logout | XML/Pilot/etc versions | Long version (with comments) | Weblog archives | Site Map | | Browse Topics

Microsoft TCP/IP benchmark speed explained

2001-08-23 15:53:55+00 by Dan Lyke 0 comments

Remember the big hullabaloo over the Mindcraft benchmark which showed NT beating Linux in a TCP/IP networking with multiple cards, allegedly because NT had a multithreaded TCP/IP stack. Mike didn't pass along a source on this as well, but he reports taht recent attempts to duplicate that benchmark failed until someone turned on the non-standard 9k MTUs in NT. Further updates as news breaks.

Speaking of this, I saw Microsoft's "99.99% uptime" claim, and I thought about the only way I've ever managed 90%+ uptimes with NT, and flashed on a new Shroedinger's Cat experiment, where you boot an NT box, turn off the monitor, and...

[ related topics: Free Software Microsoft Open Source ]

comments in ascending chronological order (reverse):