Oracle Analytics Cloud and Server

Home› Oracle Analytics› Oracle Analytics Forums› Oracle Analytics Cloud and Server› Oracle Analytics Cloud and Server

Categories

Welcome to the Oracle Analytics Community: Please complete your User Profile and upload your Profile Picture

Data Reading Best Approach

Received Response
31
Views
4
Comments
Ibrahim Azzam-186772
Ibrahim Azzam-186772 Rank 2 - Community Beginner
in Oracle Analytics Cloud and Server

Summary

Viewing the OAC data source best approach

Content

Dear Experts,

We are implementing OAC and considering the best approaches to read the data from the Oracle on-prem databases. We have a large volume of daily transactions, so what is recommended whether to read directly from the on-prem resource (example using Data Gateway) or upload the data to the cloud first ( example using ODI or datasync)?

Note: The examples are what I concluded so far without going in details of other alternatives.

Regards,

 

Answers

  • [Deleted User]
    [Deleted User] Rank 7 - Analytics Coach

    What does "large volume" mean?

    How good is your connectivity between OAC and the on-premises database? Bandwith etc.

    Have you checked how long it would take you to upload the data?

    Do you need the data real-time or would a periodical update over ODI be acceptable (so not real-time)?

    etc.

    There are literally dozens of other questions which impact such a design choice.

  • Ibrahim Azzam-186772
    Ibrahim Azzam-186772 Rank 2 - Community Beginner

    Hi Christian,

    Thanks for sharing your comments and please find my answers:

    What does "large volume" mean?

    It means a huge number of transactions will be available everyday in the on-prem database ready to be uploaded or viewed.     

    How good is your connectivity between OAC and the on-premises database? Bandwith etc.

    The connectivity in general is good and reliable.

    Do you need the data real-time or would a periodical update over ODI be acceptable (so not real-time)?

    As per the implementation scope both is needed but most the visualization will not depend on real time data.

    There are literally dozens of other questions which impact such a design choice.

    That is why I opened the post, I need initiatives....what is recommended?...what is the ideal solution used by the customers?

    Regards.

  • [Deleted User]
    [Deleted User] Rank 7 - Analytics Coach

    That still doesn't define "large" at all. Some companies may consider 10GB / day as "large". Others maybe as of 200.

    My point that I am trying to make is: It depends. There is no "one size fits all" let alone "best practice". It depends on what is best for YOU in YOUR situation with YOUR data and YOUR needs.

  • Ibrahim Azzam-186772
    Ibrahim Azzam-186772 Rank 2 - Community Beginner

    As an estimate is 10 GB per day.