So what is the performance and project success rate of node.js or Python?

Question

Python

Let's say we have a grandiose project with a three-star architecture - a lot of front-end, back-end, data storage servers.
In data warehouses there will be 3D models, pictures and other everyday trifles.
They are interested in solid answers, and not like: programmers will decide for themselves, choose what is more familiar.
Or maybe you should use several languages ​​at once? To each his own task. Hybrid - as usual with VK.
Comparison criteria:
1. The price of the issue of development.
2. Price of servers.
3. Probability of successful completion of the project.
4. Terms of development.
5. Productivity.
6. Scalability.
7. Fault tolerance.
I will also consider all other languages, if there is a practical sense.

Answer 1

In order to give a qualified and reasonable answer on all points, you need to be at least a PM with experience in such large projects in both languages ​​and technology stacks. Each such project stretches for years, the probability that PM will be lucky enough to manage large projects on both stacks during his career exists, but is not high. It is even less likely that such a PM is sitting on the Toaster on January 1st. So my opinion and the opinion of other programmers present here should be understood as the opinion of masons about the estimate for the construction of the palace. That is, all this is complete IMHO, biased towards technologies familiar from experience. However, I will try to answer as objectively as possible.
More dependent on recruiting. In general, the average pay rate for an encoder for node and python is approximately equal, slightly higher than php, but lower than java, c++ and other non-scepter languages.
Inversely proportional to productivity. Neither node nor python impose any special requirements that affect the price of servers.
It completely depends on the selection of the team and personally on the PM. The same goes for any other technology.
Now here someone will definitely mention V8 as a decisive argument in crawling node. V8, of course, is a masterpiece of its kind, but not everything is so simple, because in python (unlike js) it is not customary to write performance-critical sections on the script itself (I already wrote this in detail: What to write back-end on, in which there is a lot of mathematics? ).
Depends on the architecture of the application. Building easily scalable solutions is possible on both technologies.
It strongly depends on the architecture of the application and on the choice of tools (frameworks, libraries). Building fault-tolerant systems is possible on both technologies.

Answer 2

The main factor influencing all your questions is the experience of the team “in general” and with a specific technology in particular.
If a team has several years of experience with technology X and no experience with Y, then the first year or two when working with Y will be much more difficult. Therefore, you should choose based purely on the characteristics of the team.
From the point of view of Python vs Node, I can only say that, in my opinion, the average Python programmer is more competent than the average Node programmer, since among the latter there are many front-end developers who worked little with servers.
It is most likely impossible to answer anything more specific to the question formulated in this way.

Answer 3

superivankorolev : Node modules are written on whether they are stable or not. The language should be chosen from the tasks of the project, what do you want to do?

Answer 4

a short answer to your question
wargaming.com/ru/careers/vacancies
but the solution to the problem in general may lie outside the current answers: everything rests on personnel - the backbone of the guru is in the wings, and in such conditions it may turn out that there will be only a team in Java + PHP

Answer 5

You don't have to write anything.
All of the above is in free OO
(well, except for the asterisk, which is still under development).
Install and customize for yourself.
dev.onlyoffice.org/ru