Ignore:
Timestamp:
Dec 13, 2011, 9:24:59 PM (8 years ago)
Author:
cameron
Message:

Minor fixes.

File:
1 edited

Legend:

Unmodified
Added
Removed
  • docs/HPCA2012/final_ieee/05-corei3.tex

    r1775 r1778  
    3737This overhead has little impact on the overall performance of these parsers
    3838as they experience additional overheads related to branch mispredictions.
    39 Wne compared with Xerces and Expat, the data organization of Parabix-XML
     39When compared with Xerces and Expat, the data organization of Parabix-XML
    4040significantly reduces the overall cache miss rate; specifically, there
    4141were $7\times$ and $15\times$ fewer L1 and L2 cache misses compared to
     
    7171
    7272The high branch misprediction rate of conventional parsers is a
    73 significant overhead, with the cost of a single branch mispredic-
    74 tion on the order of 10s of CPU cycles spent to restart the processor
     73significant overhead, with the cost of a single branch misprediction
     74on the order of 10s of CPU cycles spent to restart the processor
    7575pipeline. Parabix-XML is nearly branch free and exhibits minimal
    7676dependence on the source markup density. Specifically, our study
Note: See TracChangeset for help on using the changeset viewer.