NeuroGram 2.3 is an older version of the NeuroGram system, with a client-server architecture:
- NeuroGram 2.3 server was implemented as a system service operating in COM/DCOM architecture (under Microsoft Windows®) or in CORBA® architecture (under GNU/Linux and UNIX™). Thanks to being a system service which may be permanently loaded into memory, the system’s response time can be significantly shortened. Otherwise, the system would need to load data from the dictionaries every time a connection was initiated.
- The NeuroGram 2.3 client (user’s application software) is a program written in one of the following languages: C++, Java, Python, PHP, C#. Communication with the NeuroGram 2.3 server is realized via a class specifically designed for this purpose, whose methods make Gram’s services available while covering the details of DCOM/CORBA architecture.
The user’s application software can be run on the same computer on which the NeuroGram 2.3 server is running, or on another computer in the network (in this case Gram’s server may be installed on any computer to which the user has access rights).
At one time, the NeuroGram 2.3 server can handle multiple references simultaneously – many allocated interfaces from different processes. What is more, user’s application software may open many connections with the server and use them simultaneously. Function calls in Gram’s interface are synchronous. This means that after making a function call the user waits until the function is complete and returns a result.
NeuroGram 2.3 software has been optimized both in terms of operation speed and memory usage. It involves special architectural solutions and a specialized structure of lexical data.