Is Aerohive NG a good option for my network?

  • 1
  • Question
  • Updated 2 years ago
I have been holding off on using Aerohive NG as I was told it wasn't ready for primetime just yet. This was some time ago. I want to know if I should upgrade to it now. My network is fairly basic (high school). 30 odd APs (121, 230, 390) 6 Aerohive switches (2024P), and a couple cisco switches. Wireless users authenticate using Cloudpath's Radius Server so I would need to be able to set up a walled garden to that.  2 VLANS (more when I get round to it), 4 SSID's, etc.

What say you? Thanks :)  
Photo of James Watson

James Watson

  • 70 Posts
  • 8 Reply Likes

Posted 2 years ago

  • 1
Photo of BJ

BJ, Champ

  • 374 Posts
  • 45 Reply Likes
Although I like some of the new features and layout of NG, I would hold off for awhile longer unless you need support for the new switches. I am running an on-prem production HM 6 and test NG environment in tandem.
The first observation is the learning curve, particularly if you've been a HM user for any length of time. The next item is converting your existing policies, ssids, users, cwp's, etc. That alone may be worth the holdout, especially if a conversion tool is indeed in development as rumored.
I do look forward to the future with NG. The thought of development on the API's can really bring about some cool features and applications. Anyway, there is my two cents.

Photo of Crowdie

Crowdie, Champ

  • 972 Posts
  • 272 Reply Likes
We have been holding off upgrading a number of our sites from HMOL to NG as the following features are not currently supported in NG:

  • Cloning of network policies
  • Reoccurring Private PSKs
  • Access point tags (although locations can be utilised) for classifiers

Are you using any HiveManager features that are not currently supported by NG?
Photo of Luke Harris

Luke Harris

  • 265 Posts
  • 18 Reply Likes
We have also been holding off on upgrading for a number of the reasons stated previously. The main issues for our deployment are the Access Point/Device tags and the lack of in-depth reporting features that are present within HM 6.x. 
Photo of Roberto Casula

Roberto Casula, Champ

  • 231 Posts
  • 111 Reply Likes
Just to weigh in on this. Things have moved on significantly with NG. I do think Aerohive under-estimated how long development would take, and I think they acknowledge that themselves, but we are seeing frequent incremental updates that are slowly addressing the various limitations others have highlighted. Also, the fundamentals of the NG platform architecture are very strong and Aerohive have absolutely taken the right path in what they have done there.

Having said that, we have held off onboarding NG into our portfolio and still only offer our customers HM6, though we do always put that in the context of what NG will ultimately be able to give them.

The main reason is, as others have said, that there is still a feature and functionality disparity between NG and HM6. There are simply a number of things that can be done with HM6 that cannot yet be done with NG. There are also some things which are more difficult to do in NG than was the case in HM6 that directly impact our cost to implement and to support. Some of these were mentioned by Crowdie, but there have been and continue to be others.

Of course, there are many benefits to NG as well, but from a design perspective, I can't have our presales team designing solutions around ifs and maybes regarding if and when some of the missing functionality might become available - it's a constantly moving target at the moment, and I'm wary of building on foundations that haven't yet solidified.