SMB Harwal Electric Pty Ltd is one of the foremost low voltage electric switchboard manufacturers in Australia. SMB Harwal Electric relies on its Microsoft^® Windows^® IT infrastructure to support its IT needs, ranging from running the high-tech machinery on the factory floor, to email and file sharing and printing. As the organization grows, SMB Harwal Electric needs to upgrade and improve its Microsoft^® Windows^® infrastructure to cope with the growth. In 2010, SMB Harwal Electric engaged Kiong Software to review and audit its Microsoft^® Windows^® infrastructure and then to implement a solution to improve its IT infrastructure.

The Challenge

Our client has a Microsoft^® Active Directory^® infrastructure that forms the backbone of its whole IT infrastructure. Due to the growth in the client's organization, the client has doubts that the existing infrastructure is able provide the level of performance and redundancy that its business expects and the client would like to improve the performance and resilience of its IT infrastructure. On the other hand, the client has many systems that depend on its Windows environment to function properly and any work done to improve the performance and resilience of the infrastructure must not negatively impact other systems.

Approach taken to overcome the challenge

From our review and audit of the client's Windows^® infrastructure we discovered that the client has two domain controllers in its Active Directory domain. While this setup offers some level of fault-tolerance to the IT infrastructure, we discovered that some of the critical services like DNS, DHCP and File and Print are performed by a single server and such dependencies of critical services on a single server are potential single point of failure that could bring down the whole IT environment. Furthermore, all the servers are located in the same datacentre and that is another potential single point of failure in the IT infrastructure.

We worked with our client to understand their needs and concerns and proposed a solution that will improve both the performance and reliability of its Windows infrastructure. We recommended to the client to replace an existing old server with a newer and more powerful server. In addition, we recommended distributing the critical services like DNS and DHCP across several servers. These two recommendations will improve both the performance and reliability of the Windows environment. For example, previously only one server is configured to answer DHCP requests from all the computers on the network and this puts all the burden of answering DHCP requests on one server. After the upgrade, the burden of answering the DHCP requests is shared equally between two servers, with each server responsible for answering DHCP requests from half the number of computers on the network.

 

A similar approach is used to improve the performance of the DNS service. With our recommendation, half of the computers in the network are configured to use one server as the primary DNS server and the other half is configured to use the other server as the primary DNS server.

 

For the improvement to the File and Print service, we implemented a solution to synchronize the files on the main file and print server to a standby server on regular basis so that in the event of an outage of the main file server, it is possible to redirect the users to the standby file and print server and the users will be able to continue with their work.

 

The remaining task is to improve the fault-tolerance level of the infrastructure by locating the servers in different buildings to reduce the possibility of a single point of failure.