This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Active Roles Server Web Interface - How to Implement High Availability?ARS

Hello!

So, pardon me if this has been asked in a previous post, but how does one go about ensuring high availability for a web interface? I currently have one web interface and one web database, and my business is soon expecting a huge number of users to access the website within a short time span. I was just curious what the best way is to minimize the load on the web interface.  I wasn't sure if it was to create a replicated database, or if it is to create a duplicate website....

Thank you for any help you can provide!

  • Many clients with large environments will do the following for HA:

    Several AR Admin services - all attached to a SQL Always On Availability group

    On the pure web interface side, they will implement several IIS servers hosting the web UI that are load balanced by round-robin DNS's or also commonly, behind an F5 and using a Virtual IP.  Note that you don't HAVE to co-locate the web interface and the Admin service on the same host,

    'Hope this helps.

  • Thank you so much! Is there documentation anywhere that details on how to load balance by round-robin DNS's or the F5/VIrtual IP method you described?

  • I would recommend to consider HA vs. DR scenarios, though both might overlap. Options vary and can be used together in different combinations. For example...

    (1) SQL Cluster + two ARS01,02 ADmin Servers with IIS websites (http://ARWebAdmin etc.) + LoadBalancer(internal DNS) in front

    (2) SQL Cluster in (1) can be replaced with SQL replication (legacy scenario)

    (3) some customers use VM layer for DR (in case ARS Servers got "lost"). It is more DR than HA.