Wireless Access

 View Only
  • 1.  MCR-VA-10K considerations

    Posted Oct 20, 2022 09:09 AM
    We're planning on creating a secondary AOS8 wireless system due to the capacity limitations with our existing infrastructure (we're not in a position to trial AOS10/Central and we're about to breach the 10K AP limit). Therefore, we are considering deploying a VM Conductor with on premise mobility controllers to navigate our capacity requirements, as short/medium term solution. The option of using a VM is financially more appealing and the lead times for equipment may be too long to fix our problem.

    On the face of it, reading the documentation available, there is not too much difference between running an on premise Conductor vs a VM Conductor. Is anyone aware of any subtle or major differences in the day to day operation e.g. Airgroup functionality etc. We have little to no experience running a Conductor or any kind of supporting infrastructure as a VM (within our immediate team) - so any input would be appreciated.

    The 'separate' wireless system will effectively mirror the existing system with two clusters in two data centres, but with the exception of a VM Conductor. The sites which will terminate on the new system will be geographically separate to customers on the existing system, which will avoid roaming difficulties between the two mobility domains.

    Adam


  • 2.  RE: MCR-VA-10K considerations

    Posted Oct 20, 2022 10:14 AM
    Have you looked at the memory/cpu/disk requirements of the 10K vm?  https://www.arubanetworks.com/techdocs/ArubaOS-8.x-Books/811/ArubaOS-8.11.0.0-Virtual-Appliance-Installation-Guide.pdf
    What is an on-premise VM conductor?  Do you mean local vs. at a datacenter?

    There are quite a few considerations and I suggest you either engage TAC or get professional help to make sure things are planned out efficiently and correctly.


    ------------------------------
    Any opinions expressed here are solely my own and not necessarily that of Hewlett Packard Enterprise or Aruba Networks.

    HPE Design and Deploy Guides: https://community.arubanetworks.com/support/migrated-knowledge-base?attachments=&communitykey=dcc83c62-1a3a-4dd8-94dc-92968ea6fff1&pageindex=0&pagesize=12&search=&sort=most_recent&viewtype=card
    ------------------------------



  • 3.  RE: MCR-VA-10K considerations

    Posted Oct 20, 2022 10:39 AM
    Have you looked at the memory/cpu/disk requirements of the 10K vm? 
    -Yes. I have passed this information onto our servers and storage department for review.

    What is an on-premise VM conductor?  Do you mean local vs. at a datacenter? 
    Physical mobility controllers making up two clusters, housed in diverse data centres - managed by a VM Conductor.


  • 4.  RE: MCR-VA-10K considerations

    Posted Oct 20, 2022 11:01 AM
    I think a diagram might be in order.  Your words are definitely open to  interpretation.

    ------------------------------
    Any opinions expressed here are solely my own and not necessarily that of Hewlett Packard Enterprise or Aruba Networks.

    HPE Design and Deploy Guides: https://community.arubanetworks.com/support/migrated-knowledge-base?attachments=&communitykey=dcc83c62-1a3a-4dd8-94dc-92968ea6fff1&pageindex=0&pagesize=12&search=&sort=most_recent&viewtype=card
    ------------------------------



  • 5.  RE: MCR-VA-10K considerations

    Posted Oct 21, 2022 04:57 AM
    See below a high level overview of the existing and the proposed topology.  

    The existing arrangement has in cluster and site failover. Cluster B has enough capacity in case of site failover - but does not have in cluster failover enabled (hence fewer controllers). 

    The proposed additional system plan is to mitigate the 10K AP capacity issue we currently face. The idea is to use this for geographically isolated areas/sites of the university in order to breach 10K APs (currently on around 9K). The new plan will mirror the cluster arrangements of the existing system, but on a smaller scale. 

    I hope that makes more sense.




  • 6.  RE: MCR-VA-10K considerations

    Posted Oct 21, 2022 07:57 AM
    - I don't see anything wrong with your plan, as long as from a management perspective it is manageable
    - In my opinion, I think failover/redundancy beyond the original cluster should be reserved for environments that need to be highly available, but that are prone to failures.  In many environments without an absolute need for high availability or  without frequent outages, it might create more management overhead with little benefit.
    - If you are not speaking to your Aruba\HPE Sales Engineer about this, you should be.



    ------------------------------
    Any opinions expressed here are solely my own and not necessarily that of Hewlett Packard Enterprise or Aruba Networks.

    HPE Design and Deploy Guides: https://community.arubanetworks.com/support/migrated-knowledge-base?attachments=&communitykey=dcc83c62-1a3a-4dd8-94dc-92968ea6fff1&pageindex=0&pagesize=12&search=&sort=most_recent&viewtype=card
    ------------------------------