If you are interested to reproduce the graphs, i had exactly same output 
using lmbench* lat_mem_rd *binary which estimates memory latency. 
I did that to bench different cpu before puting them in production. 

The main algorithm is to create a big array (>1g for exemple) and walk 
through an it by growing step, And each time the program return the time of 
the read. 




On Sunday, September 4, 2016 at 1:32:44 AM UTC+2, Greg Young wrote:
>
> http://www.ilikebigbits.com/blog/2014/4/21/the-myth-of-ram-part-i 
>
> without comment. 
>
> -- 
> Studying for the Turing test 
>

-- 
You received this message because you are subscribed to the Google Groups 
"mechanical-sympathy" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to mechanical-sympathy+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to