THE IT STACK
The IT Stack as I have so affectingly named it has been drawn on so many napkins and scraps of paper that at times I find it difficult to take myself seriously. Then I look with amazement on the eyes of the recipient to realize it is as relevant today as it was in the early nineties when I was first exposed to it.
Yes, sometimes, the simple is really the best approach. I am visual and therefore this imagery is even more powerful to someone like me.
To the young Information Technology leader this becomes your compass for how and why decisions will be made.
I will explain the diagram from the bottom up for that is how we provide value to deliver the why of an organization problem. This is not just some trite literary play on words. The difference between how and why is the difference between organizational success and technological blunders. The latter is all too common when the bells and whistles become more important than solving the business problem. Do not chase the technology, but rather solve the organizational problem.
Here is the stack you can use to vet your solutions and their overall placement. Learn to draw this quickly on napkins and the backs of envelopes.
This is the basic pipes and plumbing of the Information Technology infrastructure. This is the physical cabling of the network, or the wireless infrastructure that allows your computers and users to talk to each other. This is all the telecommunications lines you have going into your buildings and branch offices to extend the reach of your connected devices and users.
Watch the monthly costs from your telecom provider to ensure you are getting the best value for your dollar.
Your networks need to be stable and built like a utility. By that I mean your users must have the same faith in plugging in a computer network cable as they do plugging in a toaster. When this focus is maintained confidence will be raised in your users. Your users should NOT know what type of network you have and how it is laid out. No one cares except your team so stop trying to impress your bosses and other non-related business units with the how the ones and zeros travel between computers. I cannot stress this enough. When I go into organizations and their business leaders are exposing the details of their network both positive and negative I become deeply concerned.
Networks have to be built to be used and abused. The physical plant should be well labeled and very well organized. You would be amazed on how many problems are due to sloppy labeling and wiring connections. Take pride in NOT having a rat’s nest for a wiring closet.
This is the actual hardware that you can hug and hold and get all warm and fussy about. This has the pretty lights and fans that go whir. I get real serious about hardware and I standardize on what my team can support. Being an IBM Bigot or DELL Bigot does not help you here. This layer is now a commodity. However, if you have excellent support from a vendor then make sure it has some value on the purchase side if they are not the cheapest vendor. Having every make and model from every vendor out there does NOT help you. Buying hardware you can make on your own is great if this is in your basement and you have no friends. It does not have a place in the business world. Just because you can doesn’t mean you should. Let’s be professional here. You have much more important things to worry about.
The technology platform consists of the following elements but is not limited to these.
- Routers, switches, hubs
- Mobile devices
- Storage Arrays
- Access points
All technology equipment must be respective and supportive of the network layer. These two layers depend completely on each other. Any mismatches here can have serious repercussions.
This is where the rubber hits the road or the user interacts with the system. This should be the only piece of the technology a user can explain and get intimate and animated with. The software should fill the organizational goal. The goal may be just a simple tool for adding rows and columns (read spreadsheet) or the goal may be to have a modern and robust warehouse management system. The user interaction should be respectful of who is engaged in the process/system. A repetitive theme in my posts is “if you wouldn’t use it then why are you releasing it”.
The application must also be written to be complimentary to the technology it is sitting on. No use in building a behemoth memory intensive system if in fact the user experience can only support minimal processing capacity. Use some forethought.
Like the interdependencies between the technology layer and the network layer so is the application layer to the technology layer. There is one exception for the application layer though and that is the application should also be sensitive and supportive of the network. A final check between the application layer and the network layer is paramount. You must be aware if your application cannot support a high speed network. All too often software programmers have no understanding of how networks actually work and therefore their applications do not provide a proper user experience.
A simple list of some application layers are as follows;
- Office productivity tools (MS Office, Open Office, Lotus Smartsuite, WordPerfect)
- Operating systems
- Warehouse Management Systems (WMS)
- Transportation Management Systems (TMS)
- Enterprise Resource Planning (ERP)
- Business Intelligence
- Text Messaging
- On screen programming for your Satellite receiver
- Content Management Systems
- And many more. The list is endless
This is the actual business or organizations drivers to proceed with any effort. This is the problem that is being solved and interaction should only be with the software application layer. The organizational leadership should not know about the hardware and/or the network that is delivery the solution. This is so critical. The bottom layers should be hidden from the system users. If they know the intimate workings of your solutions then you have failed. I cannot stress this enough.
The interaction is just between the problem and the solution (application user interface). The solution may have all elements of the layers in the ultimate solutions but that is not what the organization needs to know or really should care about. Sometimes when I am in a meeting with the business units and my team is present the conversation will typically lead to one of my team member saying something like “well that will be a performance hog or man that is a lot of data” and my internal or sadly at times external response will be “so what”. That is the correct answer. Sometimes to really solve a problem we will have to update the technology, applications and the network and that is our job. That is our problem and the business users.
This is the value we bring to the organization. We bring the HOW the problem will be solved. We bring recommendations to the organization to give them the tools to succeed.
We bring value through a relevant application layer that is sitting on a stable and effective technology layer and is running on a ubiquitous reliable network layer that is responsive back to the application layer.
The decision making process is from TOP down. The organization provides the WHY and the Information Technology group responds with an effective HOW. I am a strong proponent of Information Technology providing recommendations and the organizational leadership provides decisions.
But before you give me “your shitting me, right”. There are some decisions that fall in the complete purview of Information Technology leadership. These decisions will always be based on business realities and sound business judgment. The ultimate network layer and technology layer are decisions that I do not give up without a fierce fight. However, before these decisions are made I take time to understand our budgets, long term plans, support skills etc. Once the standard is set by the Information Technology group then we only need to get approval for funding and not the actual decisions
I was once told if you can’t explain an idea on the back of an envelope or a napkin it is not worth pursuing. If this is the case, the IT stack is a sure winner in this category.