From the performance, according to published benchmark, Golang and Java, Scala probably an order of magnitude, is about 30 times Python. Second Golang compile faster, this point relative to Java, Scala has a clear advantage, once again their language characteristics determine the high Golang development efficiency, in addition because of the lack trycatch mechanism, making use of exception handling Golang development when compared to ponder more so after their relatively low maintenance costs on the line. But Golang there is currently less obvious defects third-party libraries, special libraries associated with AI.
This four languages I have used recently, favorite or golang, of course, it is very multi-slot, such as a bad package management tool (do not know the latest package management tools is not an improvement), lack of generic design (go2.0 to join?), they are everywhere if err! = nil deal (there is no try catch mechanism but instead allows developers must take care of each exception and error), so it seems that the shortcomings go into the advantage, ha ha ha, unwittingly became a blow go, go very easy in concurrent programming, with who happy, go with the coroutine and channel, I do not want to use Java multi-threaded (keyword the addition of the code and long, upsets people scalp tingling), cross-platform and cross-compiler to make very simple, did not have to deploy more to say, a new server, I do not installed, go build a binary file , does not depend on other libraries, run-time environment without the need to throw up to get away!
Go concurrent performance advantages like syntax, the Internet has a lot of people boast here that the advantage of a relatively few people mention. Go lets programmers are not familiar with the underlying smooth contact and look under the hood. Only to learn at the outset Go, master the basic syntax and programming techniques, such as avoiding traversal, avoiding data copying, etc., you can make good use of the server program up. To further the pursuit of performance, it can be dug Go source code to see its internal implementation, coupled with the tools to analyze project pprof Go to offer. Go code is very clear, analytical tools as well, at this stage you do not need to listen to people say better or bad, you have your own judgment. Go provides a feature called cgo, lets Go fast docking C / C ++ library written, you can go to become familiar with C / C ++ from cgo, compare wonderful learning path, but the learning curve is not so steep, while learning you can while output the code to see results. At that time google produce goalng of context to solve c / c ++ can do it but the wheels or less made up more difficult situation. Go election is equivalent to choose a relatively flat course, but no matter what the language or technology to do the service side, each path are sure to point to the bottom of the system, to see how to get it.
Of course, still have to do server-side development better and go to the bottom, get to know the underlying technology, the difference just before the language is hard to recruit people, does not happy with it just the problem!
FIG attached to an analogy:
