翻译文献Distributed Database Development.docVIP

大连交通大学2016届本科生毕业设计（论文）外文翻译 外文原文 Distributed Database Development Centralized databases have been in large-scale use in the data processing industry since the early 1970s.Distributed databases are very much more sophisticated than centralized databases ,and they involve complex software to handle problems such as concurrency control , deadlock prevention and the maintenance of data integrity during multi-site updates .It is only recently that these software problems have been understood ,and products which incorporate this software have only very recently started to appear on the market. The present generation of DDBs mainly involve replication of databases at several sites ,rather than distribution of individual databases. The next generation of DDBs , though ,is likely to involve distribution of data over several geographic nodes ,in order to place data as close as possible to the location at which the highest rate of activity takes place . With these databases it is likely that multi-site updates will be more commonly used than at present. This will involve very careful preparation of recovery procedures in order to ensure that the risk of loss of data integrity is driver down to an absolute minimum. In some installations it will also be necessary to carefully monitor both the network and the DDBs on an on-going basis, in order to enable the operations staff to react rapidly to any situation in which data corruption has taken place . Some work has already been done in this area, using expert systems to help in determining whether a given problem is a communications problem or a database problem. However , there is a need for these diagnostic techniques to mature to a point where they can be used on an everyday basis by computer operations staff. In the immediate future, the most intensive development work will be in the area of the optimizer. This is the module used by the DDB in order to minimize computing and communications costs are generally high compared w