Affiliation:
1. IRISA, Université de Rennes, Rennes, France
Abstract
Writing large Web applications is known to be difficult. One challenge comes from the fact that the application's logic is scattered into heterogeneous clients and servers, making it difficult to share code between both sides or to move code from one side to the other. Another challenge is performance: while Web applications rely on ever more code on the client-side, they may run on smart phones with limited hardware capabilities. These two challenges raise the following problem: how to benefit from high-level languages and libraries making code complexity easier to manage and abstracting over the clients and servers differences without trading this ease of engineering for performance? This article presents high-level abstractions defined as deep embedded DSLs in Scala that can generate efficient code leveraging the characteristics of both client and server environments. We compare performance on client-side against other candidate technologies and against hand written low-level JavaScript code. Though code written with our DSL has a high level of abstraction, our benchmark on a real world application reports that it runs as fast as hand tuned low-level JavaScript code.
Publisher
Association for Computing Machinery (ACM)
Subject
Computer Graphics and Computer-Aided Design,Software
Cited by
3 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献
1. Leveraging metamorphic testing to automatically detect inconsistencies in code generator families;Software Testing, Verification and Reliability;2019-12-20
2. Automatic non-functional testing of code generators families;ACM SIGPLAN Notices;2017-05-12
3. Automatic non-functional testing of code generators families;Proceedings of the 2016 ACM SIGPLAN International Conference on Generative Programming: Concepts and Experiences;2016-10-20