J�rg Schilling wrote:
>REAL system load is: (1 - t_old/t_new)*100 = (1 - 8.014/18.309)*100 = 56.22%
I think Jorg is overlooking the scheduling algorithm.
Like just about any other kernel, the stock Linux kernels use
a scheduler that favors processes that block a lot over those that
consume as much CPU as is available. Basically, if a process tends to
block for a while then do some computation, it will generally have a
full allocation of cpu time available to it when it is time to do its
computation, while the cpu intensive process will on average have
burned through half of its allocation at that time. So, the CPU
intensive processes will slow down disproportionately when an IO
intensive process that still has some significant CPU consumption is
run simultaneously.
Adam J. Richter __ ______________ 575 Oroville Road
[EMAIL PROTECTED] \ / Milpitas, California 95035
+1 408 309-6081 | g g d r a s i l United States of America
"Free Software For The Rest Of Us."
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
