all,
would it be too far fetched to expect a kernel RPM to set this reasonable starting limit based on the number of processers, type, and the Mhz speed of them?

You can find this out by parsing /proc/cpuinfo... Just my 2 cents.

It should be easy to come up with reasonable limits with a little testing to figure out what the "magic" equation is. A good starting point for the pentium research might be:
p iteration * 32 * (mhz / 1000)* cpu count

This would leave you with:
P90: 1 * 32* .09 * 1= 1.44 (we'll go with a 32 here)
dual PIII 800: 3 * 32 * .8 * 2 = 153 (lets round to 256 or 128)
quad xeon 2Ghz: 4 * 32 * 2 * 8 (4*dual core)= 2048

I am just a simple code monkey so go easy on me lmfao. This isn't an exact science, so there may not be a right answer.

I think it's more important to set the limit than it is for it to be perfect for your particular situation.

l8,
anon

[ reply ]

Link to this comment: http://www.securityfocus.com/comments/columns/308/31172#31172