Research Article

Transparent Runtime Migration of Loop-Based Traces of Processor Instructions to Reconfigurable Processing Units

Table 3

Communication overhead.

Kernels
Communication overhead
#Inst. of PLB CR#Inst. of FSL CRDDR-PLB (%)LMB-PLB (%)LMB-FSL (%)

count271292.456.3926.22
even_ones341892.162.2530.49
fibonacci271463.215.584.22
ham_dist351791.861.1829.98
pop_cnt32351792.161.3430.49
reverse351792.361.3430.49

compress351995.071.8239.97
divlu251092.253.7926.83
expand351995.071.7640.75
gcd321577.334.5112.81
isqrt341696.274.7745.52
maxstr251092.555.2526.89

popcount3371846.978.162.77
mpegcrc362085.547.0021.69
usqrt311889.959.5628.93

merge156.322.587.058.1417.07
merge1 (n/s)32.1715.8N/A41.9416.13
merge25722.089.670.3224.43
merge2 (n/s)3114.8N/A48.0421.26