Graphics Processing Units (GPUs) broke out by the end of 1990s, devoted to the goal of providing ubiquitous interactive 3D graphics which, a few years back then, was a far-fetched dream. By the end of decade, the technology grew exponentially, with nearly every computer containing a GPU, providing a high performance, visually rich, brilliant 3D computer graphics. This unprecedented growth was a cumulative outcome of rising demand for high-quality games, manufacturing processes advancements, and employment of inherent parallelism for computation. Today, the raw computational power of a GPU dwarfs that of the most powerful CPU, and the gap is continuously widening. World's most powerful supercomputers (e. g. , Tianhe-2(China) and Titan(USA)) use GPUs at their core [1]. This paper provides a review of GPU technology, overview of general purpose application of GPUs, architectural highlights, Enhancement of GPGPUs. In the end, we take a look at future research directions and challenges to parallel computing chips.

1. URL: http://www. top500. org/lists/2015/06/.
2. Gordon E. Moore, the co-founder and chairman emeritus of Intel and Fairchild Semiconductor.
3. URL: http://www. nvidia. com/object/tesla-servers. html.
4. NVIDIA Corporation. CUDA Toolkit 4. 0. http://developer. nvidia. com/ category/zone/cuda- zone.
5. AMD Close to the Metal (CTM). http://www. amd. com/.
6. The AMD Fusion Family of APUs. http://sites. amd. com/us/fusion/apu/ Pages/fusion. aspx.
7. URL: https://en. wikipedia. org/wiki/Sandy_Bridge.
8. URL: http://www. androidauthority. com/arm-mali-closer-look-605021/.
9. Khronos Group. OpenCL - the open standard for parallel programming on heterogeneous systems. http://www. khronos. org/opencl/
10. URL: http://allegroviva. com/gpu-computing/difference-between-gpu-and-cpu/#prettyPhoto
11. URL:http://www. nvidia. com/object/what-is-gpu-computing. html
12. 'An Introduction to Modern GPU architecture' by Ashu Rege, Director of Developer Technology, NVIDIA
13. Preliminary views on GPU technology by Burton Smith, Technical Fellow, Microsoft Formerly, Chief Scientist at Cray
14. "The Architecture and Evolution of CPU-GPU Systems for General Purpose Computing," Manish Arora, University of California, San Diego.
15. URL:http://www. nvidia. in/object/gpu-computing-in. html
16. E. Lindholm et al. NVIDIA Tesla: A unified graphics and computing ?architecture. IEEE Micro, 2008.
17. C. M. Wittenbrink et al. Fermi GF100 gpu architecture. IEEE Micro, ?2011.
18. NVIDIA's next generation cuda compute architecture: Kepler GK110. Technical report, 2012. ?
19. URL: http://devblogs. nvidia. com/parallelforall/maxwell-most-advanced-cuda-gpu-ever-made/
20. URL:http://www. karlrupp. net/2013/06/cpu-gpu-and-mic-hardware-characteristics-over-time/
21. URL:http://www. pcgamer. com/hardware-report-card-nvidia-vs-amd/
22. J. T. Adriaens et al. The case for gpgpu spatial multitasking. High Performance Computer Architecture, 2012.
23. J. Nickolls et al. The GPU computing era. IEEE Micro, 2010.
24. V. W. Lee et al. Debunking the 100x GPU vs. CPU myth: An evaluation of throughput computing on CPU and GPU. In International Symposium on Computer Architecture, 2010
25. Steven Muchnick. Advanced Compiler Design and Imple- mentation. Morgan Kaufmann, 1997
26. W. W. Fung, I. Sham, G. Yuan, and T. M. Aamodt. Dynamic Warp Formation and Scheduling for Efficient GPU Control Flow. In 40th International Symposium on Microarchitec- ture (MICRO-40), December 2007.
27. Collange, Sylvain. Stack-less SIMT Reconvergence at Low Cost, 2011. ?
28. Minsoo Rhu, Mattan Eraz, The Dual-Path Execution Model for Efficient GPU Control Flow, 2013.
29. J. Meng, D. Tarjan, and K. Skadron. Dynamic Warp Subdivision for Integrated Branch and Memory Divergence Tolerance. In 37th International Symposium on Computer Architecture (ISCA-37), 2010.
30. V. Narasiman et al. Improving GPU performance via large warps and two-level warp scheduling. In International Symposium on Microarchitecture, 2011.
31. S. W Keckler et al. GPUs and The Future of Parallel Computing. IEEE Micro,2011.
33. URL: http://www. extremetech. com/gaming/201417-nvidias-2016-roadmap-shows-huge-performance-gains-from-upcoming-pascal-architecture.

Graphics Processing Unit (gpu) Central Processing Unit (cpu) Graphics Pipeline Parallel-computing