Data storage

printable version

Our CRAFT document management system generally handles large quantities of data and images. The system uses different database engines to store all that information. This article gives an overview of these database engines.

Apart from its openness and easy connectivity, the system uses a proprietary database, which provides for the secure handling and quick accessibility of especially large quantities of data (it is designed to handle 90 to 100 million documents) and for the maximum security of the data stored. Instead of using our proprietary database management system, we can also connect the system to any SQL-based database management system, if necessary.

The Craft system stores images and data (the information stored about images) in separate databases due to the different nature of these data.

Storing images

Images are stored in a special image database, which is managed by an image database server, an integrated part of the Craft system. This database was designed for the efficient handling of large quantities of data which do not change throughout their life cycle. As a special feature, the different parts of the image database can be divided between and stored on different media, even on CDs. The image database server can even control a jukebox to access specific images stored only on CDs. The image database is protected against unauthorised data modification and unauthorised data queries. The database uses standard data formats and encryption methods, which make stored information easy to retrieve if the encryption keys are available, regardless of whether using CRAFT or any other database management system. This approach ensures long-term and system-independent accessibility of the stored data.

Storing archived data

Archived documents can be stored using the proprietary database management system of the Craft system. This proprietary database management system has the following features:

  • It was developed to handle extremely large volumes of data, thus, it can handle much larger databases using the same resources (CPU, memory);

  • It is a standard SGML/XML database, which can store SGML/XML data of any structure, exploits the advantages of this format and ensures long-term integrity of the stored data;

  • It supports the custom characteristics of the system: for instance, it allows specifying relationships between the stored data based on a number of aspects;

  • It supports storing all or part of the database on CD disks and retrieving images directly from CDs using a jukebox;

  • The system is protected against unauthorised data access and modification;

  • It stores data and search indexes separately, thus, not taking encryption into account, data are stored in SGML/XML format, i.e. as plain text.

Apart from that, archived documents can also be stored using SQL-based database management systems. The system can use the following database engines at present:

  • fast access to Oracle database engines through an OCI interface,
  • SQLite database engine with direct access,
  • access to other SQL-based database engines through an ODBC interface.

Workflow

A Workflow database is required to store and manage the data of active documents. When using such a database, it is important that the current workflow status of the data be always clearly specified. Although this task does not seem to be complicated, when you have a nationwide system which performs a large number of simultaneous operations and any of the hardware components connected to the network may go down at certain times, this is not an easy problem to solve.

The workflow storage database can be a special database engine developed by Cardinal Kft. for this purpose, or any of the SQL-based database engines already mentioned under the management of archive databases.

 
 
There are no related items.