• Àüü
  • ÀüÀÚ/Àü±â
  • Åë½Å
  • ÄÄÇ»ÅÍ
´Ý±â

»çÀÌÆ®¸Ê

Loading..

Please wait....

¿µ¹® ³í¹®Áö

Ȩ Ȩ > ¿¬±¸¹®Çå > ¿µ¹® ³í¹®Áö > Journal of EEIS

Journal of EEIS

Current Result Document :

ÇѱÛÁ¦¸ñ(Korean Title) A Formal Model for Allocation of Objects into Heterogeneous Distributed Environments
¿µ¹®Á¦¸ñ(English Title) A Formal Model for Allocation of Objects into Heterogeneous Distributed Environments
ÀúÀÚ(Author) Jaewon Oh   Seunghoon Choi   Chisu Wu  
¿ø¹®¼ö·Ïó(Citation) VOL 05 NO. 01 PP. 0041 ~ 0051 (2000. 02)
Çѱ۳»¿ë
(Korean Abstract)
¿µ¹®³»¿ë
(English Abstract)
One significant part of reengineering the object-oriented (OO) legacy systems is an effective allocation of software components in the distributed environments, to achieve performance goals, such as balancing the workloads, maximizing the degree of concurrency, and minimizing the communication costs. This paper presents a partitioning/allocation model and an allocation algorithm for mapping OO applications to heterogeneous distributed systems. Our partitioning/allocation model is different from other models in that it deals extensively with several characteristics of OO paradigm such as the inheritance, dynamic binding, frequency of method invocations, type of methods, communication cost among objects, and workload of method execution. Unlike other work in partitioning/allocation, this work includes the consideration of classes as units of partitioning/allocation to reduce a space of candidate hypotheses. Our allocation algorithm is based on the Niched Pareto Genetic Algorithm (NPGA), because such problem is a multiobjective one with non-commensurable criteria. This model and algorithm have been validated by experimenting on two typical OO systems in CORBA-based distributed environments.
Å°¿öµå(Keyword) Partitioning   Allocation   Concurrency   Load Balance   Communication Cost  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå