Implementation of Decentralized Load Sharing in Networked Workstations Using the Condor Package

نویسندگان

  • Jennifer C. Hou
  • Kang G. Shin
چکیده

In recent years a number of load sharing (LS) mechanisms have been proposed or implemented to fully utilize system resources. We have designed and implemented a decentralized real-time LS mechanism based on the Condor package 17, 18]. Two important features of our design are use of region-change broadcasts in the information policy to provide each workstation with timely state information at minimum communication cost, and use of preferred lists in the location policy to avoid task collisions. With these two features, we remove the central manager workstation in Condor, con-gure its functionalities into each participating workstation, transform Condor into a decentralized LS mechanism, and equip Condor with the capability to tolerate single workstation failures. Also discussed are the experiments on the proposed LS mechanism and the oo-the-shelf Condor package and our observations of empirical data. Any opinions, ndings, and conclusions or recommendations expressed in this paper are those of the authors and do not necessarily reeect the views of the funding agencies.

برای دانلود 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

A worldwide flock of Condors: Load sharing among workstation clusters

Condor is a distributed batch system for sharing the workload of compute-intensive jobs in a pool of Unix workstations connected by a network. In such a Condor pool, idle machines are spotted by Condor and allocated to queued jobs, thus putting otherwise unutilized capacity to e cient use. When institutions owning Condor pools cooperate, they may wish to exploit the joint capacity of their pool...

متن کامل

A Decentralized and Efficient Algorithm for Load Sharing in Networks of Workstations

This paper presents the design and evaluation of a decentralized load sharing algorithm for networks of workstations, RADIO. With respect to general distributed computing environments, networks of workstations have some peculiarities. First, the global computing power is most of the time much underutilized. Second, users of workstations need occasionally a peek of computing power. Third, workst...

متن کامل

Voltage Control and Load Sharing in a DC Islanded Microgrid Based on Disturbance Observer

Increasing DC loads along with DC nature of distributed energy resources (DERs) raises interest to DC microgrids. Conventional droop/non-droop power-sharing in microgrids suffers from load dependent voltage deviation, slow transient response, and requires the parameters of the loads, system and DERs connection status. In this paper, a new nonlinear decentralized back-stepping control strategy f...

متن کامل

Hands - On Experience with Condor , the Advanced Load

These days UNIX workstations are common in both the academia and the industry. It is very seldom the case that a computer engineer or a computer scientist does not have a UNIX workstation sitting on her/his desk. The users of those workstations often fall under one of two overlapping categories. The rst is an administrative user who works on email, paper preparation , code development and debug...

متن کامل

Reduction of Computational Complexity in Finite State Automata Explosion of Networked System Diagnosis (RESEARCH NOTE)

This research puts forward rough finite state automata which have been represented by two variants of BDD called ROBDD and ZBDD. The proposed structures have been used in networked system diagnosis and can overcome cominatorial explosion. In implementation the CUDD - Colorado University Decision Diagrams package is used. A mathematical proof for claimed complexity are provided which shows ZBDD ...

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

برای دانلود 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

عنوان ژورنال:
  • J. Parallel Distrib. Comput.

دوره 40  شماره 

صفحات  -

تاریخ انتشار 1997