论文部分内容阅读
如今,Web应用已经可以提供接近传统桌面应用的用户体验,其网页也相应地变得更加复杂,从而对Web浏览器的性能提出了巨大挑战.传统的Web浏览器通常使用单一线程处理网页,无法充分利用多处理器设备的运算能力,针对于此提出了一种并行的网页解析算法.与现有针对网页处理的并行算法不同,本算法基于数据并行的方案,通过将输入数据划分成多个部分,对其进行并行处理,再合并各个部分的结果以得到最终结果.本算法可以充分利用现有的高度优化的串行网页处理算法,并且兼容现有的Web标准和技术.在Webkit浏览器引擎上进行的实验指出,本并行算法可以有效利用多核处理器的运算能力,显著提高了网页解析过程的速度.
Nowadays, web applications can already provide a user experience close to that of traditional desktop applications, and their web pages have also become more and more complex, posing a huge challenge to the performance of web browsers, which typically use a single thread to process web pages and can not In order to make full use of computing power of multiprocessor devices, a parallel webpage parsing algorithm is proposed in this paper.Compared with the existing parallel algorithms for webpage processing, this algorithm is based on the data parallelization scheme, and by dividing the input data into multiple Part of the parallel processing, and then merge the results of the various parts to get the final result.This algorithm can take full advantage of existing highly optimized serial webpage processing algorithms, and compatible with existing Web standards and technologies.In the Webkit browser Experiments conducted on the engine indicate that the parallel algorithm can effectively utilize the computing power of the multi-core processor and significantly improve the speed of the webpage parsing process.