We will have 2 Primary Site Servers and 1 CAS. SQL will be running locally on the systems. We are running VMware vSphere 4, ESXi 4.1, So im proposing :-
Primary Site Servers and CAS
2 vCPUs
16gb RAM (due to SQL running on them)
the two primaries are due to the requirement for DR. So one primary at HQ, the other primary at DR, so if the main site gets flooded or blown up then we an run entirely from our DR site without issue.
But that won't happen automatically! You still have to re-assign all clients manually (using a script or GPO or whatever mechanism).
the two primaries are due to the requirement for DR. So one primary at HQ, the other primary at DR, so if the main site gets flooded or blown up then we an run entirely from our DR site without issue.
2 primaries and a CAS? You must be managing a lot of clients then (>100.000). Is that true? How many clients are you going to manage?
Hardware sizing is not only a matter of CPU and RAM. It also about disk I/O performance. Just have a look at "Planning for Hardware Configurations for Configuration Manager"http://technet.microsoft.com/en-us/library/hh846235.aspx).
Why have you installed a CAS and 2 primaries for managing just 700 clients? One standalone primary would be more than enough.
CPU and RAM are ok for such a low number of clients.
Why have you installed a CAS and 2 primaries for managing just 700 clients? One standalone primary would be more than enough.
CPU and RAM are ok for such a low number of clients.
I concur, a single Primary is all you need. The products designed to do this, and the design choice is a very valid one.
You are looking at a product that is extremely scalable, using just one Primary ... No need to deploy a Primary Site server to act as an administrative boundary anymore either.
the two primaries are due to the requirement for DR. So one primary at HQ, the other primary at DR, so if the main site gets flooded or blown up then we an run entirely from our DR site without issue.
the two primaries are due to the requirement for DR. So one primary at HQ, the other primary at DR, so if the main site gets flooded or blown up then we an run entirely from our DR site without issue.
But that won't happen automatically! You still have to re-assign all clients manually (using a script or GPO or whatever mechanism).
the two primaries are due to the requirement for DR. So one primary at HQ, the other primary at DR, so if the main site gets flooded or blown up then we an run entirely from our DR site without issue.
