HTTP compression

Abstract: Code-on-demand is an architectural style that a client dynamically downloads a raw script file and executes it on the client-side. This style causes a problem of network traffic because a raw script is not always compiled or minified in advance. Formatting rules, such as indents, line breaks and comments for ensuring human readability, are not necessary to the execution. In order to save wasteful data transfer, it is necessary to minify or optimize the script on the entirety of the Web. In this paper, we explore the potential for JavaScript size reduction with focus on the two reduction approaches: script minification and HTTP compression. The main two research questions are: RQ1: How many percent of websites have reduction potential? RQ2: How much JavaScript size can be reduced on the Web? Our results show that about 40% of total size of JavaScript files used on the top 500 websites can be potentially reduced by a script minification. Moreover, the current JavaScript data traffic is saving over 50% by HTTP compression. If every website was configured to use HTTP compression, we can achieve a reduction rate of 5% to 20%.

Abstract: HTTP compression is an essential tool for web speed up and network cost reduction. Not surprisingly, it is used by over 95% of top websites, saving about 75% of webpage traffic. The currently used compression format and tools were designed over 15 years ago, with static content in mind. Although the web has significantly evolved since and became highly dynamic, the compression solutions have not evolved accordingly. In the current most popular web-servers, compression effort is set as a global and static compression-level parameter. This parameter says little about the actual impact of compression on the resulting performance. Furthermore, the parameter does not take into account important dynamic factors at the server. As a result, web operators often have to blindly choose a compression level and hope for the best. In this paper we present a novel elastic compression framework that automatically sets the compression level to reach a desired working point considering the instantaneous load on the web server and the content properties. We deploy a fully-working implementation of dynamic compression in a web server, and demonstrate its benefits with experiments showing improved performance and service capacity in a variety of scenarios. Additional insights on web compression are provided by a study of the top 500 websites with respect to their compression properties and current practices.

Abstract: In this contribution, we discuss a set of techniques to improve performance of database-driven web-based applications and those techniques which are experienced on running enterprise solutions which are built on ASP.NET platform. Specifically, we will address the efficiency of three techniques: Http Compression, Ajax and database denormalization. In this effort, we show our findings of measuring performance before and after implementing specific approaches on some selected applications. We want also to disseminate the knowledge of those optimizations that can be applied on this class of applications based on our practical recommendations.

Abstract: The specification of SSL/TLS defines that data from the upper layer can be compressed in the record layer before they are encrypted. Since only the no-compression is provided for the compression algorithm in the specification, nobody can compress transmission data at the SSL/TLS protocols unless communication peers have a special agreement to use a compression algorithm. However, the compression mechanism should be useful for the users of narrower bandwidth communication lines such as ordinary analog telephone lines and N-ISDN. In order to solve this problem, the paper first introduces general-purpose compression algorithms into SSL/TLS, and shows that the average transfer times for several Japanese text files via HTTP are much improved, especially for narrow bandwidth communication lines. We also compare it with the compression mechanism on HTTP, and the resulting difference between the results of the SSL/TLS compression and HTTP compression is rather small. For improving the compression ratio, we next propose compression algorithms specialized for application protocols. As examples, we focus to HTTP and HTML, and construct the tree structures of the field names and field values. From the tree, the static dictionaries for the compression algorithms are defined. Finally, we implement the compression algorithms and show the compression ratios when several HTML files are transferred.