Monitoring and managing the cuts

Public sector IT managers can be helped to work effectively within the tighter budget, says Steve Demianyk

The timing of the recent spending review couldn’t have been worse for the many government IT departments that had already begun their regular infrastructure reviews.

There are numerous ways that the channel can help them understand their hardware and software estate so they can work with less budget. Resellers can advise public sector customers to make the most of fewer resources.

Encourage them to document their infrastructure and port-to-port connectivity. Look for layer 2/3 discovery, mapping and inventory tools that can save time and resources. A layer 2/3 discovery will discover pieces of hardware unaccounted for, or inter-device connections that no one knew were there.

The network can then be documented to simplify troubleshooting, for auditing purposes, or simply to repurpose unused resources.

At Ipswitch, we have not purchased a single piece of hardware in three years, since we moved to an internal cloud. When exploring virtualisation, the first thing to decide is which servers should be virtualised. Typically, supplementary servers like DNS, domain controllers (in small or medium size Active Directory environments), DHCP, and file and print servers are great candidates.

These types of servers generally do not take full advantage of the hardware on which they run, so are perfect targets for virtualisation. Web servers, mail servers, or small database servers can also be good. More complex servers such as application servers or database servers can also be virtualised, with careful planning.

Always analyse trends and monitor the performance of key servers that will be moving to a cloud environment, remembering to consider the peak times for any servers you plan to co-locate virtually on the same hardware. A good approach is to measure processor use, memory use, storage, network traffic volume and disk I/O. If a server is clearly overloaded to begin with, it should not be virtualised.

Do customers have multiple management consoles in place? This can severely hinder troubleshooting efforts and increase mean time to resolution (MTTR). Manually examine multiple reports and interfaces to correlate information across various types of metric.

I have found that single console management will help identify the root cause of a performance problem anywhere in the infrastructure much faster.

Customers might benefit from moving towards a solid IT management offering without all the bells and whistles of a big management framework. Open source might seem like a good option, but the apps take time to configure and can be risky when it comes to vulnerability, liability, scalability and technical support.

Public sector IT managers should be encouraged simply to look for ways to do more with less.

Steve Demianyk is channel manager at Ipswitch