GPU computing is a noun. Just like: high-performance computing.
GPU acts like an adjective. A GPU is a specific one: a countable noun. The GPU is a specific one. In this sense, one can speak of GPUs, plural: Graphical processing units.
Its main feature is [whatever]. [possessive pronoun]
It is difficult to learn. [subject pronoun: it]
The pronoun here can only be its, third person singular.
Therefore we would write:
**GPU computing [it] has become increasingly popular because of its high computing throughput.
And this sentence: GPU computing has become increasingly popular because of GPU's high computing throughput."
is incorrect. Why?
We would not repeat the noun: GPU computing, nor would we make it possessive with an apostrophe S as it has already been stated as GPU computing.
The high computing throughput refers back to the noun. Ergo, the proper pronoun is its.
The lion is a noble beast. One of its main features is its mane.
Horses are delicate animals. Their colors can be varied.
its-singular their-plural
GPU computing is the use of a GPU (graphics processing unit) as a co-processor to accelerate CPUs for general-purpose scientific and engineering computing.
The GPU accelerates applications running on the CPU by offloading some of the compute-intensive and time consuming portions of the code. The rest of the application still runs on the CPU. From a user's perspective, the application runs faster because it's using the massively parallel processing power of the GPU to boost performance. This is known as "heterogeneous" or "hybrid" computing.
Please note: GPU computing, then a GPU, which then becomes in the text: the GPU.
You move from a general notion: GPU computing, to a general idea [a GPU] to a specific one: the GPU.
If you then refer to GPU computing somewhere else in the text, the pronoun would be it as a subject pronoun and its as a possessive pronoun: Its capabilities are [whatever].
That is standard writing in English.
The text given by the OP says: the GPU's computational performance. That is correct. It is the computational performance of the GPU. Its computational performance.