BibTeX
@MISC{_douglasthain,
author = {},
title = {DOUGLAS THAIN AND MIRON LIVNY},
year = {}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
Abstra t. Distributed omputing ontinues to be an alphabet-soup of servi es and proto ols for managing omputation and storage. To live in this environment, appli ations require middleware that an transparently adapt standard interfa es to new distributed systems; su h middleware is known as an interposition agent. In this paper, we present several lessons learned about interposition agents via a progressive study of design possibilities. Although performan e is an important on ern, we pay spe ial attention to less tangible issues su h as portability, reliability, and ompatibility. We begin with a omparison of seven methods of interposition and sele t one method, the debugger trap, that is the slowest but also the most reliable. Using this method, we implement a omplete interposition agent, Parrot, that spli es existing remote I/O systems into the namespa e of standard appli ations. The primary design problem of Parrot is the mapping of xed appli ation semanti s into the semanti s of the available I/O systems. We oer a detailed dis ussion of how errors and other unexpe ted onditions must be arefully managed in order to keep this mapping inta t. We on lude with a evaluation of the performan e of the I/O proto ols employed by Parrot, and use an Andrew-like ben hmark to demonstrate that semanti dieren es have onsequen es in performan e. 1 Key words. Adaptive middleware, error diagnosis, interposition agents, virtual ma hines. 1. Introdu tion. The eld of distributed omputing has produ ed ountless systems for harnessing remote pro essors and a essing remote data. Despite the intentions of their designers, no single system has a hieved universal a eptan e or deployment. Ea h arries its own strengths and weakness in performan e, manageability, and reliability. Renewed interest in world-wide omputational systems is in reasing the number of proto ols and interfa es in play. A omplex e ology of distributed systems is here to stay. CPU /IO
Keyphrases
douglas thain miron livny interposition agent proto ols single system appli ations new distributed system tangible issue semanti dieren detailed dis ussion present several lesson servi e cpu io adapt standard interfa e virtual ma hines spe ial attention introdu tion adaptive middleware omplete interposition agent distributed system progressive study ed ountless system debugger trap spli e su middleware primary design problem andrew-like ben hmark standard appli ations essing remote data distributed omputing world-wide omputational system interfa e design possibility key word omplex ology mapping inta error diagnosis remote pro essors xed appli ation semanti
