Parallel Computing And Its Relation With GPU
Technically speaking parallel computing is using of multiple computing resources to solve a computational problem. It is also called parallel processing. So lets discuss in detail what is parallel processing and how does it work. In the initial phase a problem is broken into discrete parts that can be solved concurrently. All these are separately run multiple CPUs. Then further each part is broken down to a series of instructions. After that all the instruction from each part is executed simultaneously on different CPUs. Parallel computing is usually used in the high end computing like some tough field in science and engineering. To this question of what is parallel computing we can also say it’s example can also be seen in our mind. For example when we see any anything, our mind divides in to several components such as color, motion, shape and depth. Then it individually analyzes each one of them. It is then compared to the stored memories which help our brain to identify that particular subject. Going ahead now we can also describe the parallel computing architecture. It comprises of four main components, which are memory, control units, arithmetic & logic unit and input / output unit. The parallel computing architecture is when turned into an application, then it is called parallel computing platform. In broader sense the parallel computing architecture can be also described as parallel computing platform. This parallel computing architecture is used in many fields. One such field is graphics. When we talk of computer graphics technology, we must have heard of Nvidia.
It’s an American technology company, best known for its graphics processors. Nvidia too has developed a lot of platform parallel computing. One such is CUDA or also called Compute Unified Device Architecture. So if we are asked that what is CUDA, then we can say that it is Nvidia’s parallel computing architecture or also parallel computing platform. Its main purpose is to dynamically boost the computing performance by using the power of the graphical processing unit. The latest CUDA news reveals some interesting updates going on in CUDA. Going by the CUDA news it can be concluded that this parallel computing architecture is now used in image processing, video processing, seismic analysis, ray tracing, CT image reconstruction and much more. All those previous terms are inter-related and join to form a bigger concept. It’s called GPU. But now what is this GPU? It is a special circuit which helps in building images with the help of graphics card. Now again another question arises that what is a graphics card? It is nothing but a circuit board which generates the output for the display.