BCP Failover environment for Exchange
does anyone have any ideas on how to go about creating a failover environment for 3000 users with two datacenters?
What is BCP?
Business Continuity Planning. Its the new replacement terminology for Disaster Recovery (DR). I think DR took too much of the side of negative and the pencil pusher powers that be decided that it needed a positive image, thus coming up with something that tend to lead with 'progression' rather than just simply dealing with something under a dark cloud.
He's just basically looking for DR/HA best practices between two data centers that are hopefully off-site from one another. And I suppose the first question to ask is if his Exchange environment is currently set for DR/HA between those two data centers? If not, then do not proceed with the BCP plans for the BES environment... there would be no need.
Yes, the exchange environment is setup in two datacenters, everything is setup with failover. I need to build blackberry environment for 5,000 users with failover also. I have to build an environment that the production environment can be switched from one data center to the offsite datacener, hopefully seemlessly. What do you think? The blackberry support people weren't very helpfull, how would you go about it?
Oh, and the 5,000 isn't a typo. Since my last post the number have risen from 3000...
Thanks, I found the solution it's at Neverfailgroup.com
I was going to mention that. They had just posted here a week or so ago under another thread.
Looks interesting. Anybody use Neverfail and care to comment?
Specifically, what does it offer over and above the "out-of-the-box" solution.
There is also a product made by NSI software called DoubleTake that is very good at replicating and automatically failing over. We use this product in combination with GeoCluster to DR our Clusered Exchange 2003 Enviroment and it works great! Their website is http://www.nsisoftware.com/
Neverfail feedback (I don't use, but I'm with a MS / Neverfail SI partner)
1. Monitors application environment for things that could cause down-time (disk, memory, cpu, etc) and raises alerts
2. Users typically don't have to restart their apps - depends on the app. I have 1 project involving a web-based app that requires launching new brower session. Most Windows or mobile apps do not require users to restart.
3. Failover & switchback process is faster & easier than other products
1. Won't work in a cluster environment (i.e. replicate from primary site cluster to DR site)
2. Can't be installed on a server that's a domain controller.
Is the plan for one DC to take over for the other and vice versa? I don't think neverfail would do that. XO has some great tools for the exchange side and depending on the setup for storage EMC does as well. This isn't a simple do this and your all .. you need to figure out your timeline when things will be available and how the DC's / AD will swing over etc. Currently we need something like 50tb to replicate.
We just did our yearly test and we had exchange up and running in hour 6, BES was available 30 mins after exchange was online.
This really depends on what the business needs and to a greater degree .. what they want to spend .. $$$ :)
|All times are GMT -5. The time now is 11:49 AM.|
Powered by vBulletin® Version 3.6.12
Copyright ©2000 - 2018, Jelsoft Enterprises Ltd.