It's the wrong question to ask, since GPU is mainly being used now as data processing accelerator. Perhaps a better question is does our current CPU architectu

It's the wrong question to ask, since GPU is mainly being used now as data processing accelerator. Perhaps a better question is does our current CPU architecture is the most suitable for data processing and the likely answer is no. There are much better architecture for data processing tasks namely dataflow CPU architecture [1]. There's a new trend moving toward dataflow based architecture for data processing and AMD is one of them [2]. I think a combination of dataflow architecture optimized for array computation is the best for data processing because it generalized almost all common data processing techniques including spreadsheets, databases, graphs and matrices [3]. [1] Dataflow architecture: https://en.wikipedia.org/wiki/Dataflow_architecture https://en.wikipedia.org/wiki/Dataflow_architecture [2] AIEBLAS: Open-Source Expandable BLAS Implementation for AMD/Xilinx Versal Device: https://github.com/atlarge-research/AIE-BLAS https://github.com/atlarge-research/AIE-BLAS [3] Mathematics of Big Data: Spreadsheets, Databases, Matrices, and Graphs: https://mitpress.mit.edu/9780262038393/mathematics-of-big-data/ https://mitpress.mit.edu/9780262038393/mathematics-of-big-da...