Based On Lyapunov Optimization Of Document Research And Optimization Of Fast Conversion Scheme

The rapid development of Internet technology, our production and living all the time affected. Because it has exploded in our life, every corner and every inch of the land. More and more users are accustomed to their own private files and applications on the cloud, We found that among all of these documents, Microsoft Office documents (hereinafter referred to as the Office document) are very common, and are the status of important documents. And users to view these documents is very important, because in most cases we may just have to look on the line contents of a file without having to download. And in the cloud directly display Office documents need to be processed in the cloud can be converted into a display file format. In this paper, FlexPaper display, to avoid the problems that appear on the display.Although cloud is a powerful computing and storage capacity, to a large number of document conversion operations, mainly for the consumption of memory and CPU. So it is important to improve and optimize resource utilization. We know that the concurrent execution of multiple threads can greatly improve system throughput, better use of CPU resources. But there are difficult management, memory overflow and other problems of the conventional multi-threaded, so we introduce a multi-threaded thread pool management and monitoring.We know that the cloud environment is a cluster of servers, for the mass of requests that we need to rationalize these requests will be distributed to each server, to maximize resource utilization. Therefore, a reasonable request allocation and scheduling to improve system throughput and operational efficiency is essential. In this paper, based on Lyapunov optimization based on a wireless network to be applied to a downlink scheduling algorithm is introduced in, and its optimization and improvement. Improved processing algorithms adapted to the needs of the current cloud platforms large number of requests, improve throughput but also reduces the user’s delay.In this paper, the cloud platform for document conversion document display and multi-threaded scheduling research, to be applied to a wireless network downlink scheduling algorithm is improved and optimized. The main work is as follows:1) We through to the comparative analysis of different document transformation scheme, describes the advantages and disadvantages of different schemes, finally choose in the form of a SWF file under FlexPaper to display.2) We according to the actual situation in the cloud for application in the wireless network in a downward link scheduling algorithm was improved and optimized, increases the error feedback mechanism and error handling queue, on the basis of improving throughput guarantee user delay minimization.3) In this paper, the cloud Office document rapid transformation scheme for the design, and gives a method of document management ensures the orderly storage and management of large files in the cloud.4) Design document conversion scheme based on multi-threading, and scheduling algorithm was applied to the simple simulation experiment, verified the improved scheduling algorithm in throughput and reduce user delay.