User talk:Lionel/What kind of kernel should I make?: Difference between revisions

From OSDev.wiki
Jump to navigation Jump to search
Content added Content deleted
(Created page with "I have a problem with the notion of "monolithic == unstable". For one, stability relies very much on the quality of code. The monolithic architecture is ''statistically'' more...")
 
(→‎Reply:Solar: new section)
Line 1: Line 1:
I have a problem with the notion of "monolithic == unstable". For one, stability relies very much on the quality of code. The monolithic architecture is ''statistically'' more prone to kernel crashes than a microkernel ''if'' both have the same code quality. Two, while a microkernel might not experience ''kernel'' crashes as often, it takes quite some skill to have your system recover from, say, a crash of the filesystem server. If your system is rendered inoperable, it doesn't matter much if it was the kernel or a user-space process to blame... Hence, and to avoid rants like this very one, I would attempt to tender my statements a bit, not taking sides. -- [[User:Solar|Solar]] 04:14, 25 June 2012 (CDT)
I have a problem with the notion of "monolithic == unstable". For one, stability relies very much on the quality of code. The monolithic architecture is ''statistically'' more prone to kernel crashes than a microkernel ''if'' both have the same code quality. Two, while a microkernel might not experience ''kernel'' crashes as often, it takes quite some skill to have your system recover from, say, a crash of the filesystem server. If your system is rendered inoperable, it doesn't matter much if it was the kernel or a user-space process to blame... Hence, and to avoid rants like this very one, I would attempt to tender my statements a bit, not taking sides. -- [[User:Solar|Solar]] 04:14, 25 June 2012 (CDT)

== Reply:Solar ==

Solar, I completely agree.

Revision as of 22:57, 25 June 2012

I have a problem with the notion of "monolithic == unstable". For one, stability relies very much on the quality of code. The monolithic architecture is statistically more prone to kernel crashes than a microkernel if both have the same code quality. Two, while a microkernel might not experience kernel crashes as often, it takes quite some skill to have your system recover from, say, a crash of the filesystem server. If your system is rendered inoperable, it doesn't matter much if it was the kernel or a user-space process to blame... Hence, and to avoid rants like this very one, I would attempt to tender my statements a bit, not taking sides. -- Solar 04:14, 25 June 2012 (CDT)

Reply:Solar

Solar, I completely agree.