DocumentCode
2828823
Title
Distributed code generation using object level analysis
Author
Tiow, Tay Teng ; Yingyi, Chu ; Yang, Sun
Author_Institution
Dept. of Electr. & Comput. Eng., Nat. Univ. of Singapore, Singapore
fYear
2005
fDate
21-23 Sept. 2005
Firstpage
858
Lastpage
862
Abstract
A major issue for distributed computing systems is partitioning of the application and generating distributed code that can run correctly and efficiently on network machines. This paper introduced a new distributed code generation method using object level analysis, which is a key function of our Java Distributed code Generating and Computing platform (JDGC). Through a novel two-level object level analysis, the communication affinity metrics between runtime objects are extracted. These object-to-object affinity metrics are then used to automatically ports standard concurrent program for single machine to the form that runs on a distributed system. The contributions of this proposal are twofold. The object level analysis technique while motivated from our integrated system for object distribution, is nevertheless a general approach for extracting object-to-object communication affinity metrics and can be used in other contexts. The distributed code generation method releases the developers from any special programming considerations for distributed environment.
Keywords
Java; object-oriented programming; parallel programming; program compilers; Java; communication affinity metric; concurrent program; distributed code generation; distributed computing system; network machine; object level analysis; Application software; Computer applications; Data mining; Distributed computing; Distributed power generation; Grid computing; Information analysis; Java; Programming profession; Runtime;
fLanguage
English
Publisher
ieee
Conference_Titel
Computer and Information Technology, 2005. CIT 2005. The Fifth International Conference on
Print_ISBN
0-7695-2432-X
Type
conf
DOI
10.1109/CIT.2005.98
Filename
1562764
Link To Document