Metadata-Version: 2.1
Name: litendb
Version: 0.0.13
Summary: Big Data Analytics Toolset
Home-page: https://github.com/liten/
Author: HK Verma
Author-email: hkverma@gmail.com
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Requires-Python: >=3.12
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: numpy
Requires-Dist: cython ==3.1.4
Requires-Dist: pyarrow ==21.0.0
Requires-Dist: pandas
Requires-Dist: jupyterlab
Requires-Dist: graphviz
Requires-Dist: tbb-devel
Requires-Dist: pyspark ==3.5.3
Requires-Dist: seaborn

## Liten Database

### Concept
The Liten database constructs a tensor-based data structure that uses indices to link dimensional data with corresponding metrics. This architecture transforms traditional tabular relationships into highly efficient index lookups, resulting in dramatically faster query execution.

### Implementation Detail
A GitHub module provides a method that automatically creates data tensors during data persistence. By generating these tensors as data is stored, the system ensures query acceleration at the source, minimizing runtime computation overhead.

### Performance Benefits

Many query operations become direct index lookups rather than complex joins. Multi-dimensional tensors simplify query logic and boost performance. The approach enables high-speed analytical queries across large datasets.

### Acceleration Technology

* Novel distributed tensor-based engine with multi-dimensional hierarchical data models
* Faster execution compared to traditional relational models
* No joins, fast aggregates, and simpler query plans
* Further acceleration with GPU and FPGA hardware tensors
* Open lakehouse storage, enabling interoperability with other analytics tools
