We all know what legacy is. Its what we leave our kids when we step off the planet. We also know what load is. Its the weight of something we have to carry around with us ( not our kids that’s not fair – but perhaps the worry of them is ).
Both usually represent a burden and one that can be very tiring and stressful.
In an IT context its usually a sack of trouble and therefore, carries the title of The Legacy Load Sack.
We have been carrying it around for ages but never quite know whats in the sack, but usually experience a sack of trouble when find out.
Well inside the sack we find things we didn’t really anticipate.
People. People who have retired yet owned the intelligence and knowledge of that ‘key server or application’ that runs a business function. Didnt we notice they were getting older?
Skills. Skills that we thought were being kept up to date yet all of sudden are revealed as being at risk because we just assumed our IT people were keeping their skills current but had been working on the newer ‘sexy’ stuff. Didn’t we realise they couldn’t do both? Aren’t they heroes?
Contracts. All those service level contracts we signed years ago and unconsciously renew each year without care and attention that we either no longer need the contract in its current guise or worse the service being provided is total inappropriate and in the event of an invocation we are left with the issues and hassle.
Hardware. An obvious one but amazingly in many IT shops the legacy in a particular server or appliance manifests itself when a key application service stops working. In the autopsy review of a DR event, it is revealed that no one really owned keeping the hardware up to date, and because of a failed unsupported ‘component’ the business is really kicking the door down.
Suppliers. Yes they go out of business. When they do they often transition your support to another organization who promise the world yet deliver little. Not only does the understanding of what good service
Software. Software developers love legacy. It allows them to charge for bug fixes and feature enhancements. Yet now the decoupling of applications from hardware and the move to a lighter weight cloud stack is highlighting the interoperability issues with legacy applications that no longer confirm to good architectural practices and platform consolidation. Islands of legacy application stacks sit out there under the guise of ‘if it aint broken’, yet cause so much load to protect and babysit.
Licenses. I call these out separately because its unbelievable that many IT organizations ( and procurement departments ) have no idea of what licenses they own, the structure of the licensing agreement and any understanding of licensing benefits to upgrade and transform from their legacy estate’ Equally many licensing schemes are no longer relevant for the modern world such as those that license hardware components like CPU and concurrent usage.
Operating Systems. They go end of life. Look at Microsoft with Windows XP and Server 2003. Massive building blocks for businesses all over the world that ran very successfully, yet which are now redundant and no longer supported. Yes the promised land of cloud offers light at the end of the tunnel but today for many, end of life operating systems is a real heavy weight in the legacy load stack. Combined with suppliers removing support for end of life platforms and the security risks being magnified as vendors no longer issues security patches, the task of identifying and planning the migration path is not an easy task. Too many IT shops have ‘forgotten’ what they have and have turned a blind eye to these challenges. After all – they have other stuff to deal with.
Core infrastructure. Legacy decisions to procure networks or datacenters perhaps are the biggest legacy of all. Many years ago building your own datacenter and provisioning your own network infrastructure were bread and butter, and many very successful decisions were made and implemented. Then of course the world changed and these legacy investments suddenly made the sack even heavier. Without an easy escape clause IT organizations were constantly fighting the decision to place workloads in the cloud, while still paying a legacy load tax on their own datacenter running costs. Perhaps one of the biggest inhibitors for the cloud today.
Processes. Not one you would consider are in the sack but are firmly in there. Processes that are no longer relevant to the way workers access and consume IT create such a legacy load that the overall IT service is significantly marginalised at the cost of the business growing. For example, on boarding new users suffer from legacy decisions around process and time to deploy IT service to a new starter. Processes for backup of critical data suffer because of manual hand cranked back up routines that no longer meet the recovery time objective of the business now demanding always on IT capability. DR processes also offer legacy moments with unbelievable targets to restore business critical systems.
So a combination of any of these contribute to the ‘weight’ of the legacy load sack, and for many, this ‘weight’ is such a burden it prevents the IT organization having the spare hands to pick anything else up like IT transformation projects and other such ‘sexy’ stuff.
it is the consultant or IT manager that can understand legacy load – some have a way of measuring legacy load to keep them true – that will be the ones that truly transform their IT organization because they are the ones that can see the ‘bigger picture’ and make rationale decisions. See http://wp.me/p15xAC-xl.
The ones that don’t will carry the burden of legacy for many more years to come. Or until they actually get ‘the sack’.