- Timestamp:
- Sep 15, 2013, 2:38:51 PM (5 years ago)
- Location:
- docs/Working/re
- Files:
-
- 1 added
- 1 edited
Legend:
- Unmodified
- Added
- Removed
-
docs/Working/re/sse2.tex
r3496 r3498 7 7 8 8 9 \paragraph{GREP Implementations:} 10 We evaluate our Bitwise Data Parallel implementation against GNU grep version 2.10[?] and nrgrep version 1.12[14]. GNU grep is a popular open-source grep implementation that uses DFAs. NR-grep is one of the strongest competitors in regular expression matching performance. It uses an NFA-based approach. 11 12 \paragraph{Expressions:} 13 Each GREP implementation is tested with the five regular expressions in Table 1. Xquote matches any of the representations of a quote in xml. It is run on roads-2.gml, a 11,861,751 byte gml data file. The other four expressions are taken from Benchmark of Regex Libraries[?] and are all run on a concatenated version of the linux howto[?] which is 39,422,105 bytes in length. @ simply matches the "@" character. It demonstrates the overhead involved in matching the simplest regular expression. Date, Email, and URIOrEmail provide examples of common uses for regular expression matching. 9 14 10 15 … … 23 28 Email & \verb`([^ @]+)@([^ @]+)` \\ \hline 24 29 URIOrEmail & \verb`([a-zA-Z][a-zA-Z0-9]*)://([^ /]+)(/[^ ]*)?|([^ @]+)@([^ @]+)` \\ \hline 25 xquote & \verb`["']|"|'|�*3[49];|�*2[27];` \\ \hline30 Xquote & \verb`["']|"|'|�*3[49];|�*2[27];` \\ \hline 26 31 \end{tabular} 27 32 } … … 47 52 tick label style={font=\tiny}, 48 53 enlarge x limits=0.15, 49 enlarge y limits=0, 54 enlarge y limits={0.15, upper}, 55 ymin=0, 50 56 legend style={at={(0.5,-0.15)}, 51 57 anchor=north,legend columns=-1}, … … 54 60 bar width=7pt, 55 61 visualization depends on=y \as \rawy, 56 ymin=0,57 62 ] 58 63 \addplot[fill=black] … … 78 83 xticklabels={@,Date,Email,URIorEmail,xquote}, 79 84 tick label style={font=\tiny}, 80 enlargelimits=0.15, 85 enlarge x limits=0.15, 86 enlarge y limits={0.15, upper}, 87 ymin=0, 81 88 legend style={at={(0.5,-0.15)}, 82 89 anchor=north,legend columns=-1}, … … 85 92 bar width=7pt, 86 93 ] 87 \addplot 94 \addplot[fill=black] 88 95 file {data/instructions1.dat}; 89 \addplot 96 \addplot[fill=gray] 90 97 file {data/instructions2.dat}; 91 \addplot 98 \addplot[fill=white] 92 99 file {data/instructions3.dat}; 93 100 … … 107 114 xticklabels={@,Date,Email,URIorEmail,xquote}, 108 115 tick label style={font=\tiny}, 109 enlargelimits=0.15, 116 enlarge x limits=0.15, 117 enlarge y limits={0.15, upper}, 118 ymin=0, 110 119 legend style={at={(0.5,-0.15)}, 111 120 anchor=north,legend columns=-1}, … … 113 122 bar width=7pt, 114 123 ] 115 \addplot 124 \addplot[fill=black] 116 125 file {data/ipc1.dat}; 117 \addplot 126 \addplot[fill=gray] 118 127 file {data/ipc2.dat}; 119 \addplot 128 \addplot[fill=white] 120 129 file {data/ipc3.dat}; 121 130 … … 135 144 xticklabels={@,Date,Email,URIorEmail,xquote}, 136 145 tick label style={font=\tiny}, 137 enlargelimits=0.15, 146 enlarge x limits=0.15, 147 enlarge y limits={0.15, upper}, 148 ymin=0, 138 149 legend style={at={(0.5,-0.15)}, 139 150 anchor=north,legend columns=-1}, … … 142 153 bar width=7pt, 143 154 ] 144 \addplot 155 \addplot[fill=black] 145 156 file {data/branch-misses1.dat}; 146 \addplot 157 \addplot[fill=gray] 147 158 file {data/branch-misses2.dat}; 148 \addplot 159 \addplot[fill=white] 149 160 file {data/branch-misses3.dat}; 150 161
Note: See TracChangeset
for help on using the changeset viewer.