基于缓存和压缩的数据服务响应加速方法

Response acceleration of data services based on caching and compression

为减少数据服务调用过程中的通信开销,提出了基于缓存和压缩的数据服务响应加速方法——数据服务加速器。该方法首先采用消息模板抽取简单对象访问协议消息中有意义的对象,以减少可扩展标记语言解析的时间开销,然后通过缓存应用层的对象来减少服务执行的时间开销,同时通过使用Xmill压缩算法来减少网络传输的时间开销,通过缓存一致性维护来加强这个方法的好处。通过工业标准测试集Spring-WS-Test的实验表明,当缓存命中率达到10%时,采用DSA的数据服务的响应时间和吞吐量提高了74%和67%。

To reduce communication cost in data service calling process,the response acceleration of data services based on caching and compression was proposed.The message template was used to extract meaningful objects of Simple Object Access Protocol （SOAP）,thus the time overhead of eXtensible Markup Language （XML） was reduced.Through objects of caching application layer,the time overhead of service execution was decreased,and network transmission time was also decreased by XMill compression algorithrn.Moreover,the caching consistency was preserved to enhance the benefits.Extensive experiments based on Spring-WS-Test benchmark demonstrated that when the cache hit reached 10%,the response acceleration reduced 74％ response latency and improved 67%throughput comparing to non-caching techniques.