The link below shows some nice comparisons of different scenarios implying that none of the major nosql databases available right now completely outshines the others in terms of performance in every workload scenario. Rather, as expected, the different technologies have different strengths and weaknesses. Sort of a pick your hammer for your nail.
I feel that many organizations pick their databases (and other parts of their tech stack) not on their functional needs, but on the other operational factors. Questions like:
- How much does it cost
- How difficult is the skillset to obtain
- How difficult/expensive (same thing) is the technology to maintain
- Lame ass biases of technology managers
Of course it all comes down to cost. Of course it does. That's what we are here to do. We are here to create the best ROI for the stake holders... whatever that 'return' happens to mean. If it were all about performance, you'd by the Bugatti Veyron Super Sport rather then your crappy Toyota Corolla. The simple fact is, ROI in performance dollars is a diminishing figure. However, I don't mean to deny that the operational considerations boil down to bottom line either. It's a tightrope walk game that we play.
Increasingly however, I am finding that intelligent technologists can pick up new technologies far quicker and cheaper then they used to. Maybe it's time we start implementing solutions that are more suited to the functionality, rather then it's operationality (if that's even a word). In the long run, if organizations start to trust that our technologists can efficiently develop/maintain/support these new different technologies, if the organizations can trust their implementers maybe they could realistically realize higher ROI on both sides of the fence... Once you are finished laughing at that last statement, you may continue...
I'd like to see a study (hint hint academia) that looks at the operational vs functional cost analysis. I think that such a study is becoming more and more pertinent, especially as the years roll forward and more technologies are available. My guess is that we will see a trend toward lower cost learning newer technologies as the number of solutions increase. This may appear to fly in the face of logic, but not really. As more technologies become available, we as geeks get better at bouncing between them.
Oh, and if I unknowingly suckered you into reading this post with a poorly named title... forgive me. Here is the link that got me started thinking about all of this mumbo jumbo... Nosql performance comps:
Link to the above mentioned nosql comparisons
Get to it Academics, tell me my assumptions are BS.
No comments:
Post a Comment