A kernel level message passing service
Many of your questions may have already been answered in earlier discussions or in the FAQ. The search results page will indicate current discussions as well as past list serves, articles, and papers.
Kragen Sitaker kragen at pobox.comFri Sep 1 19:48:40 PDT 2000
- Previous message: Wireless bonding
- Next message: SCMS Released
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Mark Hahn writes: > the justification offered is that http is just another filesystem, > and since we have NFS in the kernel, why not khttpd? but putting > hhttpd in the kernel is nothing short of an admission > that Linux user-space is not viable in performance. The reason that Linux is not a microkernel OS is that running stuff split between kernel and userspace in Linux is slower than running it completely in the kernel. Is there anyone that still doubts that? You could say the same thing for almost anything that's in the kernel: the TCP/IP stack, memory management, filesystems in general, swapping. In every case, including this one, you would be essentially right, but still overstating the case. Just because context switches between the kernel and userspace are expensive does not mean that nothing should run in userspace. It just means that when you're aiming for balls-to-the-wall performance, kernel-space is a productive last-resort option. -- <kragen at pobox.com> Kragen Sitaker <http://www.pobox.com/~kragen/> Perilous to all of us are the devices of an art deeper than we ourselves possess. -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]
- Previous message: Wireless bonding
- Next message: SCMS Released
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the Beowulf mailing list
