Nowadays most of the people are focusing & talking about the trending technologies such as AI, Big Data, IoT, etc.
People are focusing on the software layer and ignoring the infrastructure although the infrastructure is main core for all of these things to operate. A missing role in the market will be coming soon as I expect will be called “Infrastructure scientist”.
It’s not a tribute for infrastructure people in the IT field but it’s a way for building the right infrastructure for workloads in a more scientific measured manner & doing like a simulation for the workloads when operating and not just a bunch of resources which will definitely work after the IaaS ( Infrastructure as a Service ) is spread in the market but in a blind way so you don’t focus on the HW used or the Virtualization technology so we didn’t get the most benefits of it depending on the IaaS that it will make it work .
Let me give you a simple example:
- If you want to run SQL Server 2012 on servers with same specs but not identical as follows
1st server specs:
CPU: Intel® Xeon® Processor E5-2699A v4
ram: 32 GB
disk type: SSD
2nd server specs:
CPU: Intel® Xeon® Processor E5-4627 v4
ram: 32 GB
disk type: SSD
Results: Both will work and maybe you won’t notice difference if 1 or 2 servers, but if you are deploying a large scale or enterprise scale deployment for big company definitely you will feel difference due to the difference in the Processors type although both have multiple cores and seems to be good processors.
But the idea that if you notice each processor specs you will find difference in performance for sure like processing latency, I/Os performance .. etc & a lot of different metrics.
That was a simple example for infrastructure scientist functions which we can have more complex examples on large scale deployments might affect performance or you won’t get the most benefits from the infrastructure you are paying for if on Cloud or you did pay for if its on-premises in your data center , or Hybrid Cloud scenarios .
If we can propagate the same idea on the rest of the machine resources like “Disks size type & configuration, Ram types & size and configuration, etc. definitely you will get different results of you are just using the hardware as it is like bunch of servers might be expensive or cheap, and for sure on the cloud you won’t be able to get all of those details.
But as much as you can gather what type of HW used within your cloud provider you can be able to achieve better results before starting your planning phase for choosing the resources of your large-scale deployment which will be addressed in the form of virtual machines types or series and its hosting region.
This role will be much more advance than “Cloud Architect” or the “Infrastructure Architect” as it won’t be only concerned with how to build the IT infrastructure and drawing Visio diagrams with beautiful icons! , as its not common that anyone focus to this level of details .
But how to build it the right way with a lot of calculations and considerations as it should be to get the maximum benefits of your resources related to what you have been paying for and not only choosing very expensive HW or choosing he highest VM series in the cloud provider to get the best results , yea for sure you might get it but it will be like a blind guessing without doing a lot of calculations that you can get the most of it .
Finally, I’m looking forward to see this role announced in the IT market soon as it will make a major difference in the IT Infrastructure field and how start building strong environments based on scientific measurements which will serve the trending technologies a lot.