source: docs/HPCA2012/final_ieee/final.aux @ 1737

Last change on this file since 1737 was 1737, checked in by ashriram, 8 years ago
File size: 10.7 KB
Line 
1\relax
2\citation{blake-isca-2010}
3\citation{esmaeilzadeh-isca-2011}
4\citation{Asanovic:EECS-2006-183}
5\citation{Cameron2008}
6\citation{xmlchip}
7\citation{CameronHerdyLin2008}
8\citation{Cameron2010}
9\citation{xmlchip,DaiNiZhu2010}
10\@writefile{toc}{\contentsline {section}{\numberline {1}Introduction}{1}}
11\citation{TR:XML}
12\citation{xerces}
13\@writefile{lof}{\contentsline {figure}{\numberline {1}{\ignorespaces XML Parser Technology Energy vs. Performance\relax }}{2}}
14\providecommand*\caption@xref[2]{\@setref\relax\@undefined{#1}}
15\newlabel{perf-energy}{{1}{2}}
16\@writefile{toc}{\contentsline {section}{\numberline {2}Background}{2}}
17\newlabel{section:background}{{2}{2}}
18\@writefile{toc}{\contentsline {subsection}{\numberline {2.1}XML}{2}}
19\@writefile{lof}{\contentsline {figure}{\numberline {2}{\ignorespaces Sample XML Document\relax }}{2}}
20\newlabel{fig:sample_xml}{{2}{2}}
21\@writefile{toc}{\contentsline {subsection}{\numberline {2.2}XML Parsers}{2}}
22\@writefile{toc}{\contentsline {section}{\numberline {3}The Parabix Framework}{3}}
23\newlabel{section:parabix}{{3}{3}}
24\@writefile{toc}{\contentsline {subsection}{\numberline {3.1}Parallel Bit Streams}{3}}
25\@writefile{lof}{\contentsline {figure}{\numberline {3}{\ignorespaces Example 7-bit ASCII Basis Bit Streams\relax }}{3}}
26\newlabel{fig:BitStreamsExample}{{3}{3}}
27\@writefile{toc}{\contentsline {subsection}{\numberline {3.2}Parabix Compilers}{4}}
28\newlabel{parabix tool chain}{{3.2}{4}}
29\@writefile{lof}{\contentsline {figure}{\numberline {5}{\ignorespaces Character Class Compiler Input/Output\relax }}{4}}
30\newlabel{fig:CCC}{{5}{4}}
31\@writefile{lof}{\contentsline {figure}{\numberline {6}{\ignorespaces Parallel Block Compiler (Pablo) Input/Output\relax }}{4}}
32\newlabel{fig:Pablo}{{6}{4}}
33\citation{xerces}
34\citation{expat}
35\citation{TR:XML}
36\@writefile{lof}{\contentsline {figure}{\numberline {4}{\ignorespaces Lexical Parsing in Parabix\relax }}{5}}
37\newlabel{fig:ParabixParsingExample}{{4}{5}}
38\@writefile{toc}{\contentsline {subsection}{\numberline {3.3}Parabix Runtime Libraries}{5}}
39\@writefile{toc}{\contentsline {section}{\numberline {4}The Parabix XML Parser}{5}}
40\newlabel{section:parser}{{4}{5}}
41\@writefile{toc}{\contentsline {section}{\numberline {5}Evaluation Framework}{5}}
42\newlabel{section:methodology}{{5}{5}}
43\newlabel{parsers}{{5}{5}}
44\citation{bellosa2001,bertran2010}
45\citation{clamp}
46\@writefile{lof}{\contentsline {figure}{\numberline {7}{\ignorespaces Parabix XML Parser Structure\relax }}{6}}
47\newlabel{parabix_arch}{{7}{6}}
48\@writefile{toc}{\contentsline {paragraph}{XML Parsers:}{6}}
49\newlabel{workloads}{{5}{6}}
50\@writefile{toc}{\contentsline {paragraph}{XML Workloads:}{6}}
51\@writefile{toc}{\contentsline {paragraph}{Platform Hardware:}{6}}
52\@writefile{toc}{\contentsline {paragraph}{Energy Measurement:}{6}}
53\@writefile{toc}{\contentsline {section}{\numberline {6}Efficiency of the Parabix-XML Parser}{6}}
54\newlabel{section:baseline}{{6}{6}}
55\@writefile{toc}{\contentsline {subsection}{\numberline {6.1}Cache behavior}{6}}
56\@writefile{lot}{\contentsline {table}{\numberline {1}{\ignorespaces XML Document Characteristics\relax }}{7}}
57\newlabel{XMLDocChars}{{1}{7}}
58\@writefile{lot}{\contentsline {table}{\numberline {2}{\ignorespaces Platform Hardware Specs\relax }}{7}}
59\newlabel{hwinfo}{{2}{7}}
60\@writefile{toc}{\contentsline {subsection}{\numberline {6.2}Branch Mispredictions}{7}}
61\newlabel{section:XML-branches}{{6.2}{7}}
62\newlabel{corei3_BR}{{9(a)}{7}}
63\newlabel{sub@corei3_BR}{{(a)}{7}}
64\newlabel{corei3_BM}{{9(b)}{7}}
65\newlabel{sub@corei3_BM}{{(b)}{7}}
66\@writefile{lof}{\contentsline {figure}{\numberline {9}{\ignorespaces Branch characteristics on the Core-i3\ per kB of input data.\relax }}{7}}
67\@writefile{lof}{\contentsline {subfigure}{\numberline{(a)}{\ignorespaces {Branch Instructions / kB}}}{7}}
68\@writefile{lof}{\contentsline {subfigure}{\numberline{(b)}{\ignorespaces {Branch Misses / kB}}}{7}}
69\@writefile{toc}{\contentsline {subsection}{\numberline {6.3}SIMD Instructions vs. Total Instructions}{7}}
70\newlabel{corei3_L1DM}{{8(a)}{8}}
71\newlabel{sub@corei3_L1DM}{{(a)}{8}}
72\newlabel{corei3_L2DM}{{8(b)}{8}}
73\newlabel{sub@corei3_L2DM}{{(b)}{8}}
74\newlabel{corei3_L3DM}{{8(c)}{8}}
75\newlabel{sub@corei3_L3DM}{{(c)}{8}}
76\@writefile{lof}{\contentsline {figure}{\numberline {8}{\ignorespaces Cache Misses per kB of input data.\relax }}{8}}
77\@writefile{lof}{\contentsline {subfigure}{\numberline{(a)}{\ignorespaces {L1 Misses}}}{8}}
78\@writefile{lof}{\contentsline {subfigure}{\numberline{(b)}{\ignorespaces {L2 Misses}}}{8}}
79\@writefile{lof}{\contentsline {subfigure}{\numberline{(c)}{\ignorespaces {L3 Misses}}}{8}}
80\newlabel{cache_misses}{{8}{8}}
81\@writefile{toc}{\contentsline {subsection}{\numberline {6.4}CPU Cycles}{8}}
82\@writefile{lof}{\contentsline {figure}{\numberline {10}{\ignorespaces SIMD Instruction Percentage\relax }}{8}}
83\newlabel{corei3_INS_p2}{{10}{8}}
84\@writefile{lof}{\contentsline {figure}{\numberline {11}{\ignorespaces Performance (CPU Cycles per kB)\relax }}{8}}
85\newlabel{corei3_TOT}{{11}{8}}
86\@writefile{toc}{\contentsline {subsection}{\numberline {6.5}Power and Energy}{8}}
87\newlabel{corei3_power}{{12(a)}{8}}
88\newlabel{sub@corei3_power}{{(a)}{8}}
89\newlabel{corei3_energy}{{12(b)}{8}}
90\newlabel{sub@corei3_energy}{{(b)}{8}}
91\@writefile{lof}{\contentsline {figure}{\numberline {12}{\ignorespaces Power profile of Parabix on Core-i3{}\relax }}{8}}
92\@writefile{lof}{\contentsline {subfigure}{\numberline{(a)}{\ignorespaces {Avg. Power (Watts)}}}{8}}
93\@writefile{lof}{\contentsline {subfigure}{\numberline{(b)}{\ignorespaces {Energy Consumption ($\mu $J per kB)}}}{8}}
94\@writefile{toc}{\contentsline {section}{\numberline {7}Evaluation of Parabix across different Hardware}{8}}
95\newlabel{section:scalability}{{7}{8}}
96\@writefile{toc}{\contentsline {subsection}{\numberline {7.1}Performance}{8}}
97\newlabel{section:scalability:intel}{{7.1}{8}}
98\newlabel{ScalabilityA}{{13(a)}{9}}
99\newlabel{sub@ScalabilityA}{{(a)}{9}}
100\newlabel{energy_Parabix2}{{13(b)}{9}}
101\newlabel{sub@energy_Parabix2}{{(b)}{9}}
102\newlabel{ScalabilityB}{{13(c)}{9}}
103\newlabel{sub@ScalabilityB}{{(c)}{9}}
104\@writefile{lof}{\contentsline {figure}{\numberline {13}{\ignorespaces Average Performance Parabix vs. Expat (y-axis: ns per kB)\relax }}{9}}
105\@writefile{lof}{\contentsline {subfigure}{\numberline{(a)}{\ignorespaces {Parabix}}}{9}}
106\@writefile{lof}{\contentsline {subfigure}{\numberline{(b)}{\ignorespaces {Avg. Energy Consumption on various hardware (nJ per kB)}}}{9}}
107\@writefile{lof}{\contentsline {subfigure}{\numberline{(c)}{\ignorespaces {Expat}}}{9}}
108\newlabel{Scalability}{{13}{9}}
109\@writefile{toc}{\contentsline {subsection}{\numberline {7.2}Parabix on Mobile processors}{9}}
110\newlabel{section:scalability:Neon{}}{{7.2}{9}}
111\newlabel{power_Parabix2}{{14(a)}{9}}
112\newlabel{sub@power_Parabix2}{{(a)}{9}}
113\@writefile{lof}{\contentsline {figure}{\numberline {14}{\ignorespaces Energy Profile of Parabix on various hardware platforms\relax }}{9}}
114\@writefile{lof}{\contentsline {subfigure}{\numberline{(a)}{\ignorespaces {Avg. Power of Parabix on various hardware (Watts)}}}{9}}
115\newlabel{arm_processing_time}{{15(a)}{10}}
116\newlabel{sub@arm_processing_time}{{(a)}{10}}
117\newlabel{relative_performance_arm}{{15(b)}{10}}
118\newlabel{sub@relative_performance_arm}{{(b)}{10}}
119\newlabel{relative_performance_intel}{{15(c)}{10}}
120\newlabel{sub@relative_performance_intel}{{(c)}{10}}
121\@writefile{lof}{\contentsline {figure}{\numberline {15}{\ignorespaces Comparison of Parabix-XML on ARM vs. Intel.\relax }}{10}}
122\@writefile{lof}{\contentsline {subfigure}{\numberline{(a)}{\ignorespaces {ARM Neon Performance (cycles per kB)}}}{10}}
123\@writefile{lof}{\contentsline {subfigure}{\numberline{(b)}{\ignorespaces {ARM Neon}}}{10}}
124\@writefile{lof}{\contentsline {subfigure}{\numberline{(c)}{\ignorespaces {Core i3}}}{10}}
125\@writefile{toc}{\contentsline {section}{\numberline {8}Parabix on AVX}{10}}
126\newlabel{section:avx}{{8}{10}}
127\@writefile{toc}{\contentsline {subsection}{\numberline {8.1}3-Operand Form}{10}}
128\@writefile{toc}{\contentsline {subsection}{\numberline {8.2}256-bit Operations}{10}}
129\@writefile{toc}{\contentsline {subsection}{\numberline {8.3}Performance Results}{10}}
130\@writefile{lof}{\contentsline {figure}{\numberline {17}{\ignorespaces Parabix Performance (y-axis: ns per kB)\relax }}{10}}
131\newlabel{avx}{{17}{10}}
132\citation{dataparallel}
133\citation{Shah:2009}
134\@writefile{lof}{\contentsline {figure}{\numberline {16}{\ignorespaces Parabix Instruction Counts (y-axis: Instructions per kB)\relax }}{11}}
135\newlabel{insmix}{{16}{11}}
136\@writefile{toc}{\contentsline {section}{\numberline {9}Multithreaded Parabix}{11}}
137\newlabel{section:multithread}{{9}{11}}
138\citation{DaiNiZhu2010}
139\citation{NicolaJohn03}
140\citation{ParaDOM2009,LiWangLiuLi2009}
141\citation{ZhangPanChiu09}
142\citation{Leventhal2009}
143\citation{DaiNiZhu2010}
144\citation{sse4}
145\citation{CameronHerdyLin2008,cameron-EuroPar2011}
146\citation{CameronLin2009}
147\citation{tan-sherwood-isca-2005}
148\@writefile{lot}{\contentsline {table}{\numberline {3}{\ignorespaces Relationship between Each Pass and Data Structures\relax }}{12}}
149\newlabel{pass_structure}{{3}{12}}
150\@writefile{lof}{\contentsline {figure}{\numberline {18}{\ignorespaces Performance (CPU Cycles per kB)\relax }}{12}}
151\newlabel{circular_buffer}{{18}{12}}
152\@writefile{toc}{\contentsline {section}{\numberline {10}Related Work}{12}}
153\newlabel{section:related}{{10}{12}}
154\@writefile{toc}{\contentsline {section}{\numberline {11}Conclusion}{12}}
155\newlabel{section:conclusion}{{11}{12}}
156\bibstyle{ieee/latex8}
157\bibdata{reference}
158\bibcite{Asanovic:EECS-2006-183}{1}
159\bibcite{bellosa2001}{2}
160\bibcite{bertran2010}{3}
161\bibcite{blake-isca-2010}{4}
162\bibcite{TR:XML}{5}
163\bibcite{Cameron2008}{6}
164\bibcite{Cameron2010}{7}
165\bibcite{cameron-EuroPar2011}{8}
166\bibcite{CameronHerdyLin2008}{9}
167\bibcite{CameronLin2009}{10}
168\bibcite{expat}{11}
169\bibcite{clamp}{12}
170\bibcite{DaiNiZhu2010}{13}
171\bibcite{esmaeilzadeh-isca-2011}{14}
172\bibcite{xerces}{15}
173\bibcite{sse4}{16}
174\bibcite{Leventhal2009}{17}
175\bibcite{xmlchip}{18}
176\bibcite{LiWangLiuLi2009}{19}
177\bibcite{dataparallel}{20}
178\bibcite{NicolaJohn03}{21}
179\bibcite{ParaDOM2009}{22}
180\bibcite{Shah:2009}{23}
181\bibcite{tan-sherwood-isca-2005}{24}
182\newlabel{pipeline_performance}{{19(a)}{13}}
183\newlabel{sub@pipeline_performance}{{(a)}{13}}
184\newlabel{pipeline_power}{{19(b)}{13}}
185\newlabel{sub@pipeline_power}{{(b)}{13}}
186\newlabel{pipeline_energy}{{19(c)}{13}}
187\newlabel{sub@pipeline_energy}{{(c)}{13}}
188\@writefile{lof}{\contentsline {figure}{\numberline {19}{\ignorespaces Multithreaded Parabix\relax }}{13}}
189\@writefile{lof}{\contentsline {subfigure}{\numberline{(a)}{\ignorespaces {Performance (Cycles / kB)}}}{13}}
190\@writefile{lof}{\contentsline {subfigure}{\numberline{(b)}{\ignorespaces {Avg. Power Consumption (Watts)}}}{13}}
191\@writefile{lof}{\contentsline {subfigure}{\numberline{(c)}{\ignorespaces {Avg. Energy Consumption (nJ / Byte)}}}{13}}
192\newlabel{multithread_perf}{{19}{13}}
193\bibcite{ZhangPanChiu09}{25}
Note: See TracBrowser for help on using the repository browser.