Sprngy implements pipelines and data lakes to provide high performance and secure data access to business as well as for traceability and auditing. Data lakes are defined to provide raw or stage data, pristine data and business data layers. Also, the storage is columnar to provide rapid access while keeping the costs low.
is a unified data and analytics platform aimed at letting users derive insights from raw data in a few clicks.
The data platform provides ability to curate and correlate large amounts of raw data with audit, balance, control and error functionality while providing traceability, quality and governance metrics.
The analytics platform provides rapid access to large amounts of curated and correlated data, rich visualizations and building analytical and predictive models for insights and foresights.
The Sprngy architecture is detailed below.
Data Platform
Sprngy uses a model-driven approach for data processing. The model driven approach is aimed at capturing:
how you would like your data to behave i.e. meta data model and
what treatment (curation, transformation, filtering, quality, governance and time travel) you would like your data to go through, in what sequence you would like your data to be processed and what insights you want your data to serve you i.e. ingestion model.
What we mean by that is that whether it is data profiling or data correlation or even importing the data from RDBMS or ingestion, curation, profiling, correlation or writing specific algorithms on the data, all of it can be done by defining models model which serve as a blueprint for the processing you want to do. These models defined how you would like your data to behave, what treatment you would like your data to go through, In what sequence you would like your data to be processed and what insights you want your data to serve you .
The model driven approach allows:
You can define the extraction, transformation and ingestion without worrying about having to code it
...
You can define
...
data cleansing and ensure your users have pristine data to work with
...
You can define
...
rules to
...
mirror data
...
for specific business uses
...
You can define advanced algorithms to provide specific insights on
...
curated data
...
For advanced users,
...
you can define machine learning models
...
to provide predictive capabilities
Sprngy has built in versioning capabilities for model driven approach to provide time travel, audit and traceability.
implements pipelines and data stores to provide high performance and secure data access to business as well as for traceability, auditing. Data lakes are defined to provide raw or stage data, pristine data and business data layers. Also, the storage is columnar to provide rapid access while keeping the costs low.