Changeset 885


Ignore:
Timestamp:
Feb 4, 2011, 11:34:44 AM (8 years ago)
Author:
cameron
Message:

Update performance results.

Location:
docs/EuroPar2011
Files:
3 edited

Legend:

Unmodified
Added
Removed
  • docs/EuroPar2011/europar-cameron.aux

    r884 r885  
    6464\bibcite{Cameron2009}{3}
    6565\bibcite{PPoPP08}{4}
    66 \bibcite{CameronHerdyLin2008}{5}
    6766\@writefile{lot}{\contentsline {table}{\numberline {2}{\ignorespaces Parser Performance (Cycles Per Byte)}}{15}}
    6867\newlabel{parsers-cpb}{{2}{15}}
     68\bibcite{CameronHerdyLin2008}{5}
    6969\bibcite{CameronLin2009}{6}
    7070\bibcite{DaiNiZhu2010}{7}
  • docs/EuroPar2011/europar-cameron.tex

    r884 r885  
    898898various \verb:xmlwf: implementations applied to the
    899899test suite.   Measurements are made on a single core of an
    900 Intel Core 2 system running a stock 64-bit Ubuntu 10.10 operating system. 
     900Intel Core 2 system running a stock 64-bit Ubuntu 10.10 operating system,
     901with all applications compiled with llvm-gcc 4.4.5 optimization level 3.
    901902Measurements are reported in CPU cycles per input byte of
    902903the XML data files in each case.
     
    910911second row.   
    911912The third row shows the best performance of our byte-at-a-time
    912 parsers, using the original  based on expat.
     913parsers, using the original  \verb:xmlwf: based on expat.
    913914
    914915The remaining rows of Table \ref{parsers-cpb} show performance
     
    924925computations performed with simulated calculation of
    925926propagated and generated carries using SIMD operations, while the
    926 ``adc'' row refers to an implementation directly employing
     927``adc64'' row refers to an implementation directly employing
    927928the processor carry flags and add-with-carry instructions on
    92892964-bit general registers.  In both cases, the overall
     
    937938\hline
    938939Parser Class & Parser & dewiki.xml  & jawiki.xml    & roads.gml  & po.xml & soap.xml  \\ \hline
    939 \multirow{3}{*}{Byte-at-a-time} & Xerces (DOM)    &    39.8    &   46.7    &  81.6    &   122.5   &    143.7  \\ \cline{2-7}
    940 & Xerces (SAX)   &     24.0   &    30.4     &  40.3    &   54.3    &   64.3     \\ \cline{2-7}
    941 & expat      &  14.2     &  17.9   &    35.4    &   44.7     &  53.6      \\ \hline
     940
     941\multirow{3}{*}{Byte-at-a-time} & Xerces (DOM)    &    37.921   &    40.559   &    72.78   &    105.497   &    125.929  \\ \cline{2-7}
     942& Xerces (SAX)   &     19.829   &    24.883   &    33.435   &    46.891   &    57.119      \\ \cline{2-7}
     943& expat      &  12.639   &    16.535   &    32.717   &    42.982   &    51.468      \\ \hline
    942944\multirow{3}{*}{Parallel Bit Stream} & Parabix1   &    8.313   &    9.335   &     13.345   &    16.136   &      19.047 \\ \cline{2-7}
    943 & gcc (simd\_add)    &  6.174   &       6.405   &       7.948   &       8.565  &        9.172 \\ \cline{2-7}
    944 & llvm (adc64)       &  5.757   &       6.142   &       6.763   &       7.424   &       7.952 \\ \hline
     945& Parabix2 (simd\_add)   &      6.103   &    6.445   &    8.034   &    8.685   &    9.53 \\ \cline{2-7}
     946& Parabix2 (adc64)       &      5.123   &    5.996   &    6.852   &    7.648   &    8.275 \\ \hline
    945947 \end{tabular}
    946948\end{center}
Note: See TracChangeset for help on using the changeset viewer.