Comparative Study on CPU, GPU and TPU
AUTHORS
P Siva Raj,Dept. of Comp. Science & Engineering, Vignan’s IIT (Autonomous), AP, India
Ch. Sekhar,Dept. of Comp. Science & Engineering, Vignan’s IIT (Autonomous), AP, India
ABSTRACT
The recent trend in the hardware technology changed as per need of the processing of huge amount of data receiving in the form of text, image, videos and graphical data. In this paper, we are going to discuss the evolutionary and need for change in hardware such as central processing units, graphical processing units and tensor processing units. Briefly explain the concept of each with architecture. Providing a comparative analysis of each unit.
KEYWORDS
CPU, GPUs, TPUs, Processing units, Tensor
REFERENCES
[1] G. Sukhdev Singh, “Comparison of single-core and multi-core processor”, IJARCSSE, vol.6, no.6, pp.423-424, (2016)
[2] Hemsoth N., First In-Depth Look at Google’s TPU Architecture, https://www.nextplatform.com/2017/04/05/ first-depth-look-googles-tpu-architecture, (2017)
[3] Geekboots, “CPU vs GPU vs TPU,” https://www.geekboots.com/story/cpu-vs-gpu-vs-tpu, (2020)
[4] N. John and D. William, “The GPU computing era”, Micro, IEEE Computer Society, vol.30, pp.56-69, (2010)
[5] M. Misic and D. Durdevic, “Evolution and trends in GPU computing”, pp.289-294, (2012)
[6] Tatourian A., “NVIDIA GPU Architecture & CUDA Programming Environment,” https://tatourian.blog /2013/09/03/nvidia-gpu-archit>ecture-cuda-programming-environment/, (2013)
[7] R. Abinash, X. Jingye, and C. Masud., “Multi-core processors: A new way forward and challenges”, ICM, pp.454-457, (2008)