Performance Evaluation of Three Dynamic Load Balancing Algorithms on Spmd Model

نویسنده

  • Najib A. Kofahi
چکیده

In this research paper, we focus on the performance of different task migration and load balancing algorithms on SPMD model based on their controlling parameters. A network of workstations has been chosen and PVM libraries have been used for implementation. Matrix multiplication has been selected as the application. Three algorithms have been investigated namely, fixed granularity, variable granularity (guided self scheduling) and one global centralized task migration algorithm. Program execution time, communication overhead and processing time have been considered as the performance measure. From the experimental results, it is found that the granularity is the controlling factor in the fixed granularity algorithm, the ratio that decreases the task size in guided self scheduling plays a vital role and that if the parameters are chosen carefully, variable task granularity and the task migration algorithms perform better than the fixed granularity.

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Empirical Study of Variable Granularity and Global Centralized Load Balancing Algorithms

Task migration and load sharing algorithms are two load balancing strategies that are essential in distributed memory multiprocessor as well as in multicomputer environments. Dynamic load balancing is more suitable in heterogeneous systems. Various load sharing and global centralized algorithms have been proposed in the literature. These algorithms demand careful investigation about their suita...

متن کامل

A Scalable Dynamic Load-Balancing Algorithm for SPMD Applications on a Non-Dedicated Heterogeneous Network of Workstations (HNOW)

Network based distributed computing has become a cheap alternative to dedicated parallel supercomputer systems due to the recent advances in high-speed network. These systems are becoming widely available in academic and industrial environments. To benefit from the maximum computation power of these systems, it is necessary to use all available resources, namely old machines in addition to more...

متن کامل

Dynamic Load Balancing in a Message Passing Virtual Parallel Machine Dynamic Load Balancing in a Message Passing Virtual Parallel Machine

In this paper we will look into the problem of dynamic balancing of tasks in a het-erogenous parallel computing environment. The parallel programs are assumed to be executed in the Single Program Multiple Data (SPMD) style. The criteria for re-balancing the load are discussed; the eeect of data movement required in the load balancing is considered; and novel algorithms of dynamic load balancing...

متن کامل

Dynamic load balancing of distributed SPMD computations with explicit message-passing

Distributed systems have the potentiality of becoming an alternative platform for parallel computations. However, there are still many obstacles to overcome, one of the most serious is that distributed systems typically consist of shared heterogeneous components with highly variable computational power. In this paper we present a load balancing support that checks the load status and, if necess...

متن کامل

A framework for SPMD applications with load balancing

This work describes SAMBA, a framework for the development of parallel SPMD applications with load balancing. SAMBA contains the structure common to different SPMD applications and a library of loadbalancing algorithms. This structure allows the developer to concentrate on the specific problem at hand. The load-balancing library makes easier the identification of the most appropriate strategy f...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2012