Data Visualisation for IT - ORDVIT Ltd
Dr Chris Tofts – Founder ORDVIT Ltd.
‘…. first remove the plank from your own eye’
The combination of phenomenal amounts of computing power, fast networks to transfer and join data from multiple sources, along with high resolution display devices has allowed us to manipulate, combine and display information in ways that are simply astounding! You can easily vanish for days into the D3 gallery, the webGL experiments or the Aframe pages, just wandering around the modern wonders of data displayed in each of them. All of these frameworks are highly dependent on deep IT support stacks of software, hardware and comms – and yet almost none of these examples (logical web maps not withstanding) are of IT systems themselves.
The complexity and scale of modern IT systems should be frankly frightening, presenting many major vital comprehension challenges that must be supported by effective data and model visualisation systems. In the wider social economic context, management of IT leads to the creation of significant volumes of data, sometimes the wrong data but that is for another article. But how do the teams of humans creating this data comprehend it? From all too painful personal experience largely we use spreadsheets with the occasional pivot table and, if produced by an ‘expert’, maybe a bar graph or even a Gantt chart. There are many elements of the IT world that create large challenging data sets:
- Business requirements and solution scope/approach
- Contract creation – interaction of legal and technical aspects
- Solution design – no! box and line diagrams are simply not enough
- Solution implementation planning – the dreaded project Gantt chart
- Transformation plans – how are these represented?
- Solution monitoring during delivery
- Solution updates and improvements
Each of these areas generates significant amounts of highly complex interdependent data, let alone the whole of the data interactions from an end to end perspective. Getting this information wrong, or even just inconsistent within, or between, project stages is one of the major causes of the repeated observation of large scale IT project failures – and that’s the ones that get made public!
At the root all of these problems is the requirement that human decision makers (yes really even in IT) can comprehend and respond to complex information – and we still present this information as text, spreadsheets and static graphs. Its almost as if we have not managed to move on from the printing press and flattened pulped trees as the dominant mode of data delivery. Even now I’m typing into a presentation system that simply tries to reflect the behaviour of an inked type stamp hitting a pulped tree – is this really the limit of our current capabilities?
Within ORDVIT we believe that the data that represents complex IT projects should be treated with the same care and affection as any political, social or economically impactful data set, and as a result have investigated data visualisations for activities as diverse as:
- Drill downs into complex costing models
- Illustrating independence and dependencies in high availability systems
- Server stacking modelling: load balancing over time
- Illustrating variation in response time of complex delivery systems: micro-service architectures
- Server interactions in large data centres
- Customer interactions with services
- User interactions with Data Bases
- Evolution of server support systems over time
- Representing supplier impact in complex projects – there’s never just one supplier really!
- Service credit models
- SLA alignment in service provision
All of these presentations make use of ‘modern’ data visualisation techniques and are presented dynamically within a browser, rather than statically on a tree!
We have the capability to improve our understanding all aspects of the complex IT systems that are commissioned, improved, migrated and impact pretty much all of our society. Engendering better comprehension of complex IT will result in the reduction the number of failures within each of these activities – and consequently the full cost of IT ‘improvements’. If we do not invest in understanding the IT we buy, build and use then we risk the failure of everything which depends on these systems - which currently is pretty much everything modern humans rely on.
For more examples of Data Visualisation for IT see ORDVIT
