Title
A Trace-Based JIT Compilation Framework for XQuery
Abstract
Query is a functional language for querying XML data. In network environment, queries are generated dynamically and executed in interpretation way. This means query programs cannot be compiled in advance to achieve a high efficiency. The just-in-time (JIT) compliation technique base on method or trace have been propsed to utilize the compilation technique. Since queries are usually short and lack of user-defined functions, we think the trace-based JIT compilation is more applicable than the method-based compilation. In this paper, we propose a trace-based JIT compilation framework to improve the performance of Query execution. Program branches executed frequently are identified as traces and compiled into object codes. We also improve the trace detection technique. Interconnected traces are merged into a trace tree and compiled at runtime to avoid unnecessary overhead of execution environment alternation between interpretation and compilation. Experiment results show our trace-based JIT compilation is more efficient than interpretation and method-based JIT compilation.
Year
DOI
Venue
2014
10.1109/ICECCS.2014.30
ICECCS
Keywords
Field
DocType
xml data querying,functional languages,object codes,functional language,trace-based jit compilation framework,compilation technique,just-in-time compliation technique,query languages,trace-based compilation,query execution,xquery, jit compilation, trace-based compilation, secd abstract machine,xml,runtime compilation,secd abstract machine,trace tree,program diagnostics,method-based compilation,trace detection technique,interconnected traces,xquery,network environment,query programs,jit compilation,program compilers,user-defined functions,radiation detectors,engines,java
Native Image Generator,Dynamic compilation,Programming language,Functional programming,Computer science,Xml data,JIT spraying,Real-time computing,Just-in-time compilation,Java,Database,XQuery
Conference
Citations 
PageRank 
References 
0
0.34
7
Authors
4
Name
Order
Citations
PageRank
Chenzhi Wu100.34
Husheng Liao22011.82
Chenglong Yu300.34
Hang Su4107.98