3 Replies Latest reply on Nov 4, 2010 12:27 PM by metalhead

    Agent Handler confusion

      I've been reading through the Agent Handler (AH) documentation, when and why they should be used however I'm still a little confused.

      For this new deployment we are definitely looking at the benefits that the AH provides for failover and future scalability.

       

      Here is the scenario -

      • currently 1000 user client base and 50+ managed servers
      • 5 geographically dispersed sites
      • smallest number of users at one site is 100
      • slowest WAN link is 8 Mbps
      • deployment of the Total Endpoint Protection or Endpoint Protection Advanced suite
      • deployment of Groupwise for Exchange
      • product suite very likely to extend in the future

       

      We were planning an AH deployment in each site with a distributed repository. Some concerns around this -

      • Is this overkill?
      • Should the AH role be deployed to a server already performing a function (i.e. file/print)?
      • Is the WAN speed sufficient? (doco suggests - each agent handler needs a relatively high speed, low latency connection to the database)
      • Does an AH take the place of a Super Agent in terms of agent wake-up calls?

       

      Thanks in advance for any responses.

       

       

      Message was edited by: jerry28 on 03/11/10 19:23:44 CDT