摘要
随着大数据技术的发展,多线程高并发等技术已经越来越成为大数据处理中的关键技术。非阻塞式I/O(new I/O,NIO)技术作为一种分布式高并发技术被广泛应用,但对于大数据量的通信往往需要很多的时间才能完成。Google提出的Protocol Buffer序列化压缩技术相对于传统序列化效率高、时间短、使用简单。文章将传统NIO技术与Protocol Buffer相结合,在分布式系统不同节点通信中,极大地降低了分布式系统的网络负载,大大节省了数据传输时间。
With the development of big data technology,multi-threaded high concurrency distributed system has become the main trend in large data processing.NIO as a distributed high concurrency technology is widely used,but for the data of large scale communication often requires a lot of time to complete.The Protocol Buffer serialization compression technology proposed by Google compared with traditional methods has the advantages of more high-efficiency,faster and easier to use.The article combines traditional NIO and Protocol Buffer to implement communications in distributed environment,which reduces the load of network and saves the data transfer time greatly.
出处
《微型机与应用》
2016年第13期19-20,24,共3页
Microcomputer & Its Applications
