This preview shows page 1. Sign up to view the full content.
Unformatted text preview: 1: mountain: A program that generates the memory mountain. The mountain program calls the run function with different working set sizes and strides. Working set sizes start at 1 KB, increasing by a factor of two, to a maximum of 8 MB. Strides range from 1 to 16. For each combination of working set size and stride, mountain prints the read throughout, in units of MB/s. The mhz function in line 19 (not shown) is a system-dependent routine that estimates the CPU clock frequency, using techniques described in Chapter 9. 6.6. PUTTING IT TOGETHER: THE IMPACT OF CACHES ON PROGRAM PERFORMANCE 329 Every computer has a unique memory mountain that characterizes the capabilities of its memory system. For example, Figure 6.42 shows the memory mountain for an Intel Pentium III Xeon system.
1200 read throughput (MB/s) 1000 L1
800 Pentium III Xeon 550 MHz 16 KB on-chip L1 d-cache 16 KB on-chip L1 i-cache 512 KB off-chip unified L2 cache 600 400 xe Slopes of Spatial Locality L2
200 Ridges of Temporal Locality 0 s1 s3 8k s5 2k working set size (bytes) mem
s7 s9 512...
View Full Document