LeoBase
New technologies
All LeoBase’s advantages are based on our own unique technologies and its implementation is not a simple improvement of existing algorithms, but the development of new solutions that can influence the process of development of information technologies related to searching, storing and processing data. Most of them may be applied not only in DBMS, but in a wide range of other fields. First of all, let’s consider these technologies.
Some of these technologies were patented, apllications on other technologies are being prepared for filing. Unfortunately, we cannot demonstrate our technologies, because we have only an out-of-date 16-bit version of DBMS LeoBase.
The implementation of 32-bit version of LeoBase will be based on a successful experience of implementation of 16-bit version and improved algorithms.
This section describe a number of technologies that will be integrated in LeoBase. To get more information on a technology, use the corresponding link.
 | Large integer data arrays storage and processing (LIDA). The mechanism of selections and selected records is based on this technology in LeoBase. Due to this technology and the structure of indices LeoBase has not problem with query optimization. The result of any search operation is a large integer data array. The ability to combine several LIDA allows LeoBase to process any search query without processing records one by one, which results in a gain in speed many times in comparison with existing DBMS. | ||||
| All data in LeoBase (including indices and unstructured data are stored in the so-called ‘containers’. Containers automatically cache information and decrease the number of disk operations. Due to data storage containers DBMS LeoBase uses more efficiently memory for caching data. For example, if 5% of memory from the size of all databases is available, the effect from data caching will be rather noticeable. Containers are indispensable while working with BLOB (binary large objects). | ||||
| The technologies of Smart bookmarks and combined document will make it possible to integrate additional features into DBMS. Combined document is a document that consists of several fragments of one or more documents. While modifying a source document, all changes will be displayed in the combined document. This ability is indispensable while making reports structuring big documents. | ||||
| Method for indexing data and the structure of indices. It will allow increasing the speed of index search. Most of existing systems are mainly based on B-tree indices. These indices can function properly only with static data and from time to time it is necessary to optimize them. The given method solves these problems. This method was tested in practice and more than 2000 clients used systems based on the given indices. This technology can be integrated in DBMS and any information systems. | ||||
| Method
for indexing text contents of documents and performing phrase search | ||||
| Method of searching for documents similar in text and meaning. This technology integrates a number of inventions:
|
It should be noted that the technology of searching for similar documents is rather unique and differs considerably from conventional phrase searching.
Information
from different sources is added to archive stored on a user’s computer.
This information often duplicates previously saved data and a user has to
process it again, for example to view it. This problem is especially important
when several users fill simultaneously an archive with document of one subject
matter in multi-user operating mode. The technology of document retrieval with
similar contents allows not only to avoid information duplication, but also to
optimize retrieval and processing of documents with similar subject matter among
already saved ones in archives. This technology permits to classify
automatically documents added to archives. The comments of the user who added
them to archives allow other users to save time for information processing.
While searching for documents with similar contents, the whole set of unique
words appeared in a document, their word-forms and semantic analogs will be
taken into account. This function is not implemented properly in most of search
systems because of enormous losses of time required for processing complex
queries. The use of the given technology is indispensable for work with
information retrieval systems, e-mail programs and other warehouses of data in
different formats
The technology of searching for similar documents can be integrated in any product.
Conclusion
While
describing LeoBase, we touched only basic technologies. In fact the total amount
of new solutions implemented in LeoBase exceeds considerably mentioned ones.
Moreover, the integration of all technologies will produce a greater affect than
their simple arithmetic addition.
In addition to mentioned technologies, LeoBase has its own original data storage structure, which is also our know-how and it has a great effect on the system operating speed. In our opinion, at the moment relational DBMS exhausted all their capabilities and they cannon manage an increased volume of unstructured information. Therefore, our post-relational DBMS is very perspective and if you invest in it, you will win.
|
|
| Home | Dms | Internet office | Software rental center | Database Leobase | Investment | Russian |
| Copyright 2002. SoftInform company. Contact us: leobase@softinform.com |
|
|