classable is a basic feature provided by the Laxcus Big Data management system that transforms a class into a string of byte arrays, or reverses a byte array into a class. This feature is very similar to the serialization provided by Java (Serializable), but the difference is that it can be defined by the user, including the selection of data, the style of the data, the data structure, and a series of rules.
The advantage is that we get rid of the type of Java serialization that is rigidly defined by the system, and we can freely organize the data we need, including some potentially private data, and the inconvenience is that because of this freedom, programmers need to make sacrifices to write code to implement this technology.
In spite of this, the Laxcus Big Data management system has done the job to the extreme Jane. API provides a class interface, the user needs to do is to inherit this interface in their own class, interface only two methods:
<1> int Build (classwriter writer);
<2> int Resolve (Classreader reader);
The first method is to output the class to a class-writable writer (classwriter), which writes data in all formats that are currently known.
The second method is to reverse the write operation and restore the data from a class-readable reader (Classreader) to a class.
The design of the Laxcus is because, in the previous large data cluster, is responsible for organizing the implementation of big data computing middleware computing components, using the Java system default serialization to produce data. The amount of data generated by serialization is very large, which is responsible for network communication, network computation and data storage. The more critical problem is that, in serialized data, there is a lot of data that is not needed in the calculation process, and they become "chicken ribs" in the data calculation of big data.
The current Laxcus middleware computing components, after the transformation of the Laxcus, the data in the large data cluster is significantly reduced, the general reduction in the amount of about 4-5 times, the individual has more than 10 times times. The speed of data transmission in the network, as well as the ability of computer to deal with the data, has been improved obviously.
The ability to class can be very effective in boosting big data capabilities.?
The classable of Big data calculation
