The need to make IT Lean

http://www.bbc.co.uk/radio4/news/inbusiness/inbusiness_20080131.shtml

I was listening to ‘In Business’, it’s a radio show on BBC Radio 4, there’s a related programme on the BBC World Service. Anyway, it was talking about ‘lean business’ which as I understand it, is a derivative a practice based on the way Toyota do business. It focuses on being lean through changing the way you do business, for example, by analyzing your helpdesk calls you can see what applications, what systems are causing the most calls to be logged, so that investment can be obtained to reduce their support cost. It’s an interesting concept, I wonder how we might achieve this in IT, what changes would occur in the IT infrastructure to reduce support costs?

Off the top of my head, the following occurred to me:

• Desktop and laptop computers refreshed every 18months – reduce hardware support cost, avoid ever getting tied to a desktop, avoid finding the desktop doesn’t meet the new operating system requirements, prevent the operating system getting unhealthy to the point where a rebuild is required.
• Replace the servers every two years to reduce the hardware support contract cost, to make the server a fixed operational cost, and benefit from more energy efficient power supplies/components.
• Perform a defragmentation and scheduled reboot of all desktops to maintain their system health ready for use when users come into work. Let’s not forget that the desktop being up for six months might be great from an end user point of view, but rebooting this will improve system performance and prevent those system hangs where an application causes the system to freeze, or the user account gets locked out because the user’s password has changed etc.
• Standardize on a set of development tools and platforms, this might comprise of for example Java and .NET, but reduce the range of layered tools, programming languages and components needed to run the application code on the server or desktop.
• Bring the development of application under IT production – IT develops and supports the application code so that we can link the strategy, availability is key as is adequate logging for troubleshooting for fault analysis.
• Virtualize the infrastructure and applications – everything should operate using a grid, web, Citrix or cloud scenario, where the actual processing and workload is separated from the end user for compliance and availability – that one server goes down should not take down the risk calculation application.
• IT to be accountable for lost revenue in the event of an application or infrastructure failure – this requires IT to be able to decide on the solution put in place, the change processes and systems put in place to manage that service. Bring to an end all those desktops or legacy ‘high maintenance’ systems which perform critical functions for an application.