Hello!
1. On the real machine, is there any way to get the frequency of L1 or L2 cache on windows OS?
2. where does the latency of accessing memory comes from? I think the cpu cycles it spent is calculated by the memory frequnecy and the CPU frequency. Is that right? but what about the Front Side Bus?
Thank you very much!
The latency of memory accesses (L1 and L2) is depended on which CPU is being used.
Please go to Intel Customer Support email form, http://supportmail.intel.com/scripts-emf/welcome.aspx
to get more informatino on your question.
1. For all modern CPUs L1 and L2 run at the same frequency as the CPU core.
2. Latency of memory access as observed from the CPU is calculated in CPU clock cycles. Unless you are accessing uncached memory or you have highly irregular access pattern most of the time you will not observe true memory latency.
Finally, latency of memory access as observed from the memory side depends mostly on the CAS signal (Column Address Strobe) but other memory timings have their role too.