source: docs/PACT2011/02-background.tex @ 949

Last change on this file since 949 was 949, checked in by lindanl, 9 years ago

more outline on paper

File size: 1.6 KB
Line 
1\section{Background}
2\label{section:background}
3XML documents tend to be verbose---especially in the case of SOAP and WSDL. Processing these documents typically requires parsing them from a text-based format into an application-specific one.
4Cameron et al.'s work in \cite{CameronHerdyLin2008} shows that both parser selection and markup density have a substantial impact in the computational cost of processing XML documents.
5
6
7In order to determine how and which performance factors influence energy consumption,
8we intend to use the Fluke i410 current clamp in conjunction with PMCs to compare the per parser invocation and per source XML byte energy usage of three XML parsers:
9Expat 2.0.1, Xerces-C++ 3.1.1 (SAX2), and Parabix2. All three parsers are C/C++ based, event-driven, stream-oriented XML parsers.
10
11
12
13Think the person reading doesn't know much about XML parsers.
14
15
16Need gory details on byte at a time parsers. Pictures.
17Xerces. Explain overall dataflow and control flow for these parsers.
18Briefly highlight inefficiencies.
19
20
21
22Talk about the usage of SIMD instructions and how it might help. Lead
23onto briefly describe the key technology behind parabix.
24
25
26
27
28
29
30
31
32
33
34The first two parsers employ traditional byte-at-a-time methods of
35parsing; these parsers were selected based on their popularity in the
36marketplace and the availability of source code for deeper analysis.
37The Fluke i410 current clamp is a digital multimeter that reads the
38magnetic field of a live electrical cable to determine the current
39passing through it without affecting the underlying hardware.
40
41
42\textbf{No Need to talk about resource usage until  later}
43
Note: See TracBrowser for help on using the repository browser.