FRAXSES

FUNCTIONALITY

CONNECT

IS IT DIFFICULT TO GET TO YOUR DATA?

Fraxses has over 230 connectors to allow you to connect to your data quickly and easily.

To access your data sources, you will need the correct credentials and security to allow Fraxses to connect with speed and ease.

DISCOVER

HOW DO I KNOW WHAT IS INSIDE A DATA SOURCE?

Once you have connected to a data source, Fraxses will start the discovery process. This involves reading the metadata from the data source. You will then be able to see all the entities and attributes, together with their types. Fraxses also records all the data entity relationships during this process.

This is also true for Parquet, CSV and Excel files – fraXses will show the attribute type depending on the contents of a column.

So, what happens if you connect to an unstructured data source or a columnar data source? The platform will discover the contents and still display it in a structured and relational format for ease of use.

MANIPULATE

I NEED TO REPORT ON DATA FROM DIFFERENT DATA SOURCES AND I.T. IS BUSY

As a low-code/no-code platform, Fraxses requires data knowledge rather than development skills.

The platform allows you to build Data Objects, which are virtual tables comprising the data you want to see. These objects use the base tables that were discovered and enable you to create joins between data from different sources and even different database technologies. As long as there is a data connection, you can join tables and select which attributes you'd like to see.

The Data Object Builder also allows you to add virtual fields and calculations as needed. This includes filtering and conditional statements, as well as a calculation builder. Again, no coding is needed.

Further to this, the Data Object Builder enables you to build objects on top of objects to ‘n’-level deep. By creating base objects and building on top of them, you can reuse any object in the hierarchy in a different object. The idea is to create objects that can be reused by different levels / departments in your company.

If you have tried and tested calculations, these can be added in SQL-99 format – so there's no need to re-engineer what you have available.

AUDIT

WHERE DO THE VALUES THAT ARE PRESENTED TO YOU COME FROM?

Once you have built your Data Objects, their structure can be viewed in our Pedigree visual, which also features data lineage functionality.

It shows you, in a hierarchical fashion, exactly where the base data originates from, how it has been manipulated and what fields have been used. It also shows which values were calculated and the computation that was used.

You can then very quickly determine that the values presented are from the source / field you would expect them to be. This process can also be used to ‘sign off’ an object – meaning that everyone in the organisation agrees to the way that the data is selected and calculated.

What you have now created is a validated object, with all the IP captured in the object, and the logic visible to whoever needs to see it. This means that IP does not reside in the logic of a report, Stored Procedure, ETL or a very complicated spreadsheet. Thus, if you have built a ‘Gross Profit Object’, then everyone in the business can use it and get the exact same result without external manipulation.

EXPOSE

HOW CAN I ADD EXTRA INFORMATION THAT IS NOT IN THE DATABASE?

Firstly, you can connect to our logical database and treat it like any other relational database.

Connection to a Data Object is as easy as:

  • A REST API call – a Data Object is, by default a REST API
  • Connecting Excel, via ODBC/JDBC, to the  database
  • Using any chosen reporting tool and connecting it to the database

The fact that no code has to be written in any reporting tool is a significant time-saving factor. In addition, the data is always ‘fresh’ and everyone sees the exact same values. You can literally expose data within half an hour of discovering your data source.

ENRICH

HOW CAN I EASILY ALLOW PEOPLE INSIDE AND OUTSIDE THE BUSINESS TO GET CONTROLLED ACCESS TO DATA?

The platform allows you to add external data i.e. spreadsheets or CSV files. These inputs are treated just like any database data source and therefore you can use additional values to enrich the data in your database.

You can also connect to external data sources (API’s, external databases) in real time to combine internal and external data. This works well for enriching data in your organisation.

STREAMING/DATA PIPELINE

CAN I INGEST AND BLEND EXTERNAL DATA?

Part of the enrichment of your data could also be an external stream, a Twitter feed, Stock Prices or a Reuters stream.

This data can be persisted in the Fraxses platform or in any other database of your choice. Analytics can then be done on these saved values or transformations made and values added before saving this information for later use.

You might also need to move data for a faster query speed or to take the load off a production system.

The Flowz module offers:

  • A low-code/no-code data pipeline using Apache Spark (most cloud pipelines require coding in steps);
  • A pipeline that is fully integrated into the Fraxses metadata stack and microservices;
  • The option to leverage Data Objects in pipeline;
  • A unified platform for batch and streaming jobs;
  • The ability to join streams to streams and batches in the same pipeline;
  • A no staging event - everything is managed in one Flowz execution (In-memory);
  • An intelligent result caching for multiple steps to use results it has a dependency on; and
  • The option to scale from thousands to billions of records.

SECURITY

HOW DO I ENSURE THAT PEOPLE ARE ONLY ALLOWED TO SEE CERTAIN DATA?

All access in the Fraxses platform is role driven. From access to data sources and Data Objects, to the lowest granular level of data rows and columns. Not only can datasets be filtered according to user level, data can also be masked depending on role level access.

As an example, one user might be able to see a phone number, while for security reasons, a user with a different role sees asterisks with only the last 4 digits exposed. 

This gives you, the company, total control over who can see what without having to make any coding changes or writing different views for different users.

MULTI-KEY DISCOVERY

HOW DO I COMBINE DIFFERENT DATA SOURCES WHEN I DON’T KNOW WHAT THE LINKS ARE?

Data only makes sense when it can be related. The relationship between datasets is usually (but not always) available in a single database system, and almost never available across databases.

In a federated environment, it is the starting point to understand your data landscape e.g. the customer in my Oracle database is the same customer as in my SQL server database.

Fraxses has:

  • An advanced algorithm that discovers relationships in hours and days instead of weeks or months;
  • The ability to discover relationships between data sources that your business did not know it had;
  • The capability to discover partial relationships and errors between related data;
  • An intuitive interface that shows the relationship confidence for easy approval;
  • The option to use sampling or execute over all data;
  • The ability to easily add a new dataset and discover additional relationships.