Javelin: Internet‐based parallel computing using Java

Abstract

SUMMARYJava offers the basic infrastructure needed to integrate computers connected to the Internet into a seamless distributed computational resource: an infrastructure for running coarse-grained parallel applications on numerous, anonymous machines. First, we sketch such a resource's essential technical properties. Then, we present Javelin, an infrastructure for global computing. The system is based on Internet software that is interoperable, increasingly secure and ubiquitous: Java-enabled Web technology. Ease of participation is seen as a key property for such a resource to realize the vision of a multiprocessing environment comprising thousands of computers. Javelin's architecture and implementation require participants to have access to only a Java-enabled Web browser. The security constraints implied by this, the resulting architecture and the current implementation are presented. The Javelin architecture is intended to be a substrate on which various programming models may be implemented. Different computation servers, each corresponding to a different programming model, are supported by Javelin's architecture. Several are presented briefly. Experimental results are given in the form of micro-benchmarks, a Mersenne Prime application and a ray-tracing application that run on a heterogeneous network of several parallel machines, workstations and PCs.

Browser-based Harnessing of Voluntary Computational Power

“…We identify common problems 3 http://www.internetworldstats.com/stats.htm Therefore, the ideal jobs for BBVC should be easily divisible into small, computationally independent tasks -or, at least, if the computation runs in phases, tasks should be computationally autonomous within each phase. The general problem categories considered by researchers working on BBVC were called piecework computations [16], master-worker [26], coarse-grained [22], malleable applications [19] or bags of tasks [23].…”

“…During our survey we encountered the following problems identified as well-suited for BBVC: evolutionary/genetic algorithms [23][25] [44][58] [64], simulated annealing [23] [64], colony optimization [23], particle swarm optimization [23], artificial bee colonies [23], brute force search [64], cryptography [9][64] [80], game-tree evaluation [64], mapreduce problems with computation-intensive map phase [22] [47][50] [57][67] [79], montecarlo simulations [19][54] [55] branch-and-bound [16] and its limited subset, bulksynchronous-parallel (BSP) [68], and, finally, the algorithms based on directed acyclic graphs [19].…”

“…Raytracing was very popular, starting with one of the first applet-based approach (distributed applet-based massive parallel processing, DAMPP) from 1997 [77]. Other work which suggested raytracing as example problem are [16][23] [71], though the author of [71] admitted his test results were not encouraging.…”

“…[19]. Other problems used in experiments were: DES code breaking [10], matrix multiplication [16] [68], Traveling Salesman Problem [16][24] [67], Mandelbrot generation [69] [70], searching for Mersenne primes using LucasLehmer primality test [56] [29], Collatz hypothesis verification [56], the factorization of large integer into two primes and Pearson correlation evaluation on set of genetic samples [9].…”

“…For those problems linear or even superlinear speedup [16][57] was often achieved, with experiments employing from 80 [67] through 200 [47] up to 400 clients [22]. A number of researchers [23][47] [54][55] [57][71] noticed the problem of "speedup decay": adding more workers caused speedup to level off or even drop.…”