基于Storm的选股回测与计算系统的设计与实现
发布时间：2019-05-31 15:47
【摘要】：随着中国经济的腾飞,越来越多的民众参与到了股票市场中。广大股民们有着自己的投资理念,他们希望可以预测自己的投资策略在未来的表现。虽然未来是不可知的,但中国的股票市场已有近30年的历史,投资策略在股票市场历史上的表现可以为投资策略提供数据支持。选股回测与计算系统为股民提供了自定义选股问句和策略信息,并在历史区间内模拟交易和计算收益情况的功能,指导股民投资。本系统是同行业中第一个提供此类功能的产品,并且具有消耗资源少、可靠性高、计算快的特点。本文系统采用了新的分布式实时计算框架Storm,充分利用Storm的并行计算能力,将复杂计算任务切分成多个子任务,提高计算速度。本文以mongoDB作为主要的数据存储载体,将MyBatis作为数据库持久层框架,并采用Spring架构。在此基础上,本文分析了系统的项目背景和技术背景,重点介绍需求分析和系统设计,详细阐述了重要模块的实现细节。在测试环境下,本文系统可以在500ms内完成复杂的运算。目前,系统已经被成功部署和运行在真实环境中,运行情况良好。
[Abstract]:With the rapid development of China's economy, more and more people have participated in the stock market. The majority of investors have their own investment ideas, they hope to be able to predict the performance of their investment strategy in the future. Although the future is unknowable, China's stock market has a history of nearly 30 years. The performance of investment strategies in the history of the stock market can provide data support for investment strategies. The stock selection return test and calculation system provides investors with custom stock selection questions and strategy information, and simulates the function of trading and calculating returns in the historical range to guide the investors to invest. This system is the first product in the same industry to provide this kind of function, and has the characteristics of low consumption of resources, high reliability and fast calculation. In this paper, a new distributed real-time computing framework Storm, is adopted to make full use of the parallel computing ability of Storm, and the complex computing tasks are divided into several subtasks to improve the computing speed. In this paper, mongoDB is used as the main data storage carrier, MyBatis is used as the database persistence layer framework, and Spring architecture is adopted. On this basis, this paper analyzes the project background and technical background of the system, focuses on the requirements analysis and system design, and expounds the implementation details of the important modules in detail. In the test environment, the system can complete the complex operation in 500ms. At present, the system has been successfully deployed and run in the real environment, running well.
【学位授予单位】：南京大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TP311.52
本文编号：2489823
[Abstract]:With the rapid development of China's economy, more and more people have participated in the stock market. The majority of investors have their own investment ideas, they hope to be able to predict the performance of their investment strategy in the future. Although the future is unknowable, China's stock market has a history of nearly 30 years. The performance of investment strategies in the history of the stock market can provide data support for investment strategies. The stock selection return test and calculation system provides investors with custom stock selection questions and strategy information, and simulates the function of trading and calculating returns in the historical range to guide the investors to invest. This system is the first product in the same industry to provide this kind of function, and has the characteristics of low consumption of resources, high reliability and fast calculation. In this paper, a new distributed real-time computing framework Storm, is adopted to make full use of the parallel computing ability of Storm, and the complex computing tasks are divided into several subtasks to improve the computing speed. In this paper, mongoDB is used as the main data storage carrier, MyBatis is used as the database persistence layer framework, and Spring architecture is adopted. On this basis, this paper analyzes the project background and technical background of the system, focuses on the requirements analysis and system design, and expounds the implementation details of the important modules in detail. In the test environment, the system can complete the complex operation in 500ms. At present, the system has been successfully deployed and run in the real environment, running well.
【学位授予单位】：南京大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TP311.52
【参考文献】
相关硕士学位论文 前5条
1 李爽;基于Spark的数据处理分析系统的设计与实现[D];北京交通大学;2015年
2 龙少杭;基于Storm的实时大数据分析系统的研究与实现[D];上海交通大学;2015年
3 戴菲;基于Storm的实时计算系统的研究与实现[D];西安电子科技大学;2014年
4 江民彬;非关系型与关系型空间数据库对比分析与协同应用研究[D];首都师范大学;2013年
5 顾昕;分布式流式计算框架关键技术的研究与实现[D];北京邮电大学;2012年
，本文编号：2489823
本文链接：https://www.wllwen.com/kejilunwen/ruanjiangongchenglunwen/2489823.html
