Parallel and Distributed Web Programming with Actors

JavaScript is the predominant language when it comes to developing applications for the web. Many of today’s web-based systems are implemented solely in JavaScript due to its applicability to both client and server-side development. Programmers have an ever-growing need to express parallelism due to the computationally intensive nature of modern web applications. Currently, JavaScript tries to satisfy this need through actor-based parallelism constructs. However, we argue that these constructs suffer from design flaws which hamper programmers to elegantly express parallelism in web applications. In this paper we present Spiders.ts, a unified actor framework to express both parallelism and distribution. In Spiders.ts, programmers can easily specify the coarse-grained parallelism needs of modern web applications. Moreover, Spiders.ts’ built-in distribution features allow programmers to express client-to-server, server-to-server and client-to-client communication simply by using actors. Through benchmarks we show that our framework is able to substantially improve the performance of web applications. Moreover, we demonstrate its expressive power by comparing implementations of a distributed case study application.