Design, build and migrate Core Windows Services for new data centre to allow closure of an old Data Centre

BP, London

The Challenge

CHALLENGE: The lease was due on our clients data centre premises which would not be renewed.   Relocate Active Directory, DNS, WINS and DHCP to services to the new Data Centre location.  We were tasked with producing the logical design and migration plan for the relocation of these services and therefore the services for the end users

REMIT: Introduce the new Core Services model and swing the traffic of dependent services to the new data centre with zero impact/outage to the business.

Our Solution

Taking advantage of latest (at the time) Windows Server version and Active Directory Functional level we introduced a new concept coined the AD Cube theory into the clients Data Centre.  This new Active Directory site structure allowed controlled distribution of specific AD services depending on the AD servrice required (replication traffic or end user traffic or top level FSMO role type traffic.  It also allowed the removal of Domain Controllers from remote sites (network qulaity dependent).  DNS was designed to 'marry' the new AD design separating DNS requests traffic for windows clients from DNS request traffic from Application Servers and from DNS request traffic from Core Infrastructure Servers.  DHCP servers were strategically placed within the User Layers of the new AD model, again in-line with controlled placement and allocation of services.  WINS, although a legacy services was still heavily utilised by older systems so the existing WINS replication model was augmented, optimised, more inline with the new AD global replication topology.

The Result

The implementation of the new design was completed on time and crucially before legacy data centre lease expired and significant late exit fees were applied to the client.  The new resilient Core Infrastructure design was a class leading success ensuring control of traffic load against critical infrastructure yet allowing automatic fail-over of services in the event of a failure of planned maintenance.  The removal of Core Services Services out of the remote sites and serving them form the data centre allowed the client to decommission a significant number of servers reducing their infrastructure footprint and TCO.

Start a project. Get in touch