FRAXSES SPRINGBOK RELEASE WEBINAR
Download PDF

INTENDA BACKGROUND

SOLUTIONS OFFERED:

CUSTOMER BASE:

FRAXSES DATA FABRIC

END-TO-END DATA FABRIC PLATFORM COMPONENTS:

  • Data Federation and Virtualization
  • Data Lineage and Pedigree
  • Data Pipelining and IoT
  • External data source blending
  • Reporting Data Caching (New Age data warehouse / Intelligent Lake)
  • Data Relationship Discovery
  • Configured ETL and Migration
  • Write-back services to data sources
  • Data Visualizer
  • Unstructured data analysis
  • App configurator releasing end Q2 2021
  • Underpinned by a microservice architecture / Methodology for your company

SPRINGBOK RELEASE

ARCHITECTURE

COMPONENT ARCHITECTURE

BUILT ON THE SHOULDERS OF GIANTS

CONSISTENT CODE
EXECUTION
ENVIRONMENT

SIMPLIFIED
DEPLOYMENT
ENVIRONMENT

USER
AUTHENTICATED
IMAGE STORE

FUTURE PROOF EXTENSIBILITY

LANGUAGE AGNOSTIC

Components can be written in different languages

SEAMLESS COORDINATION

Smart wrappers used by the events engine provide seamless coordination, communications and logging

SERVERLESS-MICROSERVICES

The old microservices could communicate with third party components but they needed to be a JSON service. The serverless components will allow the platform development team to write components in the language that best suits the job: Core: Scala, Rust, C# and Haskell Data science: python, julia, F# and RML & AI: python, Haskell, Clojure and prolog or any other variation.

This functionality can be extended for client usage so they can write components for their instance of fraXses in the languages their development / data science teams understand which can then be tightly integrated into the fraXses platform.

We have connectors for data sources and in a similar way we will have connectors for languages.

EVENTS ENGINE

The microservices events engine has been upgraded to work with the new serverless environment.


The coordinator component orchestrates a stream of events by communicating with the smart wrapper of each component.

SMART WRAPPER

The smart wrapper provides the ability to run serverless code in isolation with unified communications, encryption and logging in the component language.

SECURITY SERVER

Security server to enforce data access restrictions based on rules defined by users/groups.

  • There are 4 different types of restrictions
    • Data masking
    • Row filters
    • Column filters
    • Table access


  • Keep the same result set structure
    • Clients or 3rd party applications don’t need to change their code (Column filtering)

  • Scalable

  • Low overhead
 

DATA ACCESS CONTROL OVER FEDERATED DATA

WRITE BACK

  • FETCH META

  • DATA OBJECT VALIDATION

  • GENERATE SQL FOR TARGET DATABASE

  • AUDIT TRANSACTION

FIRST MULTI LANGUAGE EVENT

WRITE BACK

  • DATA OBJECT CREATED USING LEGOZ

  • FLAG YOUR FIELDS TO SAVE

  • WRITEBACK WILL DO THE REST

COMPLEX OBJECT EXAMPLE

POINT SOLUTION IN NUMBERS

  • Data source discovery involved 70 tables, some containing more than 100 million records. Fraxses took less than a minute to discover and capture the initial metadata.

  • The multi-key discovery process took 74 hours, and eradicated the annual data discovery overhead of 10,000 man-hours.

  • Discovering 2 Oracle data sources and both DB2 data sources involved moving 629 tables containing over 10,000 columns.

  • 3 terabytes of data containing 24 billion records were moved into Fraxses’ columnar storage (Parquet files). This took 12 hours using 6 cores – with 32 cores, it would have been completed in half the time.

  • Initial single key relationship discovery was completed in under 4 hours. The process would have taken an estimated 52 hours using a single core rather than a cluster of servers.

  • The Domestic and International Operations’ primary data sources were connected in 2 days.

  • Fraxses enabled data visualization to be completed in minutes, and executed queries on record sets containing more than 140 million records in 2 to 3 seconds.

  • Key relationship discovery: 1 key took 10 minutes, 2 key took 2 hours.

  • Data objects were configured and tested in 2 days.

OUTCOME

The Fraxses implementation was conducted over 14 days, and presented the bank with an excellent ROI:

  • The 360°customer view that had been lacking was successfully delivered.

  • Self-service: Client managers were shown how to access the data using the bank’s preferred BI tool. They were also shown how to perform validations and fine tuning on data objects.

  • For the first time, client managers were able to conduct real-time analysis on any corporate client’s account across the entire bank, and promptly provide their clients with up-to-the[1]minute reports.

  • The ability to rapidly deliver comprehensive analysis, regardless of underlying systems, represented a distinct advantage in the highly competitive international banking arena.

  • Instead of spending 10,000 man-hours on data discovery each year and incurring costs of €2,000,000 in the process, the Fraxses implementation enabled the bank to dedicate just one resource to data discovery on a part-time basis, and at negligible cost.
If you're interested in running a Fraxses point solution in your organization, contact Intenda today.
Contact Us
If you're interested in running a Fraxses point solution in your organization, contact Intenda today.
Contact Us