3GPP Benchmark Summary

This file details a summary of how the numbers were obtained for this benchmark.

Benchmark Description lines of code # of constructs in the program Number of filters in the expanded graph
filters pipelines splitjoins feedbackloops
3GPP 3GPP Radio Access Protocol 1087 16 10 18 0 48
Benchmark 250 MHz RAW processor C on a 2.2 GHz Intel Pentium IV
StreamIt on 16 tiles C on a single tile
Utilization # of tiles used MFLOPS Throughput (per 105 cycles) Throughput (per 105 cycles) Throughput (per 105 cycles)
3GPP 18% 16 44 119.6 17.3 65.7

StreamIt Number Calculations

No StreamIt numbers were available.

C Code Single Tile Raw Number Calculations

first iteration done at 0x6397 (25495 cycles).
second iteration done at 0xbdcb (48587 cycles) (delta=23092)
third iteration done at 0x117ff (71679 cycles) (delta=23092)
Based on these cycle counts, each iteration takes 23092 first/23092 second cycles (average = 23092).

4 outputs every 23092 cycles normalized to 10^5 cycles, 4*(100000/23092) = 17.32202 outputs / 10^5 cycles.

flops reported are 979 flops, which is (979 flops/23092 cycles) * 250 million cycles/second = 10.598909 MFLOPS.

Utilization numbers reported were 362267 useful cycles / 369472 total cycles = 0.9804992

C Code Number Calculations

Performing 10 million iterations at 4 outputs/iteration.
Runtime for 10^7 iterations is 112.09 seconds.

Number of cycles per iteration: 10^7 iterations/ 112.09 second * 4 outputs / 1 iteration * 1 second / 2.2*10^9 cycles * 10^5 cycles = 16.2207 outputs / 10^5 cycles.