The internet giant implemented a compact algorithm in four languages – C++, Java, Scala and its own programming language Go – and then benchmarked results to find "factors of difference".
"We find that in terms of performance, C++ wins out by a large margin," the paper says.
However, despite C++ trumping the others in performance, Google also suggests that it requires the "most extensive tuning efforts, many of which were done at a level of sophistication that would not be available to the average programmer".
This compares to Java, which the paper describes as the "simplest to implement", but also the hardest to analyse for performance.
The research found that Scala had "powerful language features", which allowed for the "best optimisation of code complexity".
Google also ran tests on Go, a language it began developing in 2007. In May 2010, the company said that Go was being used for "real stuff", by which it probably meant distributed web services.
However, Go did not stand up too well against the more established languages, as the tests indicated that the compilers for it are "still immature", which was reflected in "both performance and binary size".
By eliminating high entry costs for big data analysis, you can convert more raw data into valuable business insight.
A discussion of the "risk perception gap", its implications and how it can be closed