So I read this blog on system capacity planning yesterday. I actually had to read it twice to make sure I did not miss something…because when I got to the end, I really and truly did not think the author had said a single thing. At least it did not fulfill the promise of the title.
The piece was ostensibly on planning the amount of space one would need for a particular system, just the files to be stored, and it boiled down to this:
- Sum up all the files you plan to store in the system
- Allocate enough space for what you plan to store
Let me be clear, item 1 is find out the amount of storage you are already using. It’s like in the classic “Princess Bride”, where one is left thinking, “you keep using that word ‘planning’, I do not think it means what you think it means”. Planning calls to mind the idea of considering what might occur in, you know, the FUTURE.
I guarantee you will have a problem if you follow this advice, just as soon as one of your users creates even a single new file. BOOM. You’ve reached capacity.
Perhaps to add to the feeling that this blog was somehow more substantial, or just as a bonus, there was a mathematical formula to go along with it (paraphrased below):
If you have 500 50mb files of type x, and 1000 25mb files of type y, the formula is as follows: (#of x files)*(size of x files)+(#of y files)*(size of y files) = Amount of space required.
Now I grant that there might be someone somewhere to whom this would be useful, but I think anyone calling themselves a systems engineer, on any level, would find this to be utterly useless. It’s like someone telling you that, for a 25 mile trip in your car, you are going to need gas enough to go 25 miles. Ok…thanks?
If you haven’t been able to figure this much out on your own, maybe systems engineering is not for you. Or driving. Or blogging.