VLIW+SIMD: Decoding the Parallel Computing Genes of AI Chips

VLIW+SIMD: Decoding the Parallel Computing Genes of AI Chips

“IT有得聊”是机械工业出版社旗下IT专业资讯和服务平台,致力于帮助读者在广义的IT领域里,掌握更专业、更实用的知识与技能,快速提升职场竞争力。 点击蓝色微信名可快速关注我们! AI processors need to handle a large amount of parallel data, which requires a high degree of parallelism. There are various strategies for increasing the parallelism of processors, with common approaches including enhancing Instruction Level Parallelism (ILP), Data Level Parallelism (DLP), and Thread Level Parallelism (TLP). Enhancing Instruction Level Parallelism refers to increasing … Read more

Faster And More Accurate Data Computation With Next-Gen DSP

Faster And More Accurate Data Computation With Next-Gen DSP

The architecture of traditional Digital Signal Processors (DSP) has become inadequate for specific signal processing computational applications. However, the combination of Very Long Instruction Word (VLIW) and Single Instruction/Multi Data (SIMD) architectures can provide the parallel throughput required for high computational performance, with data typically being 16, 24, and 32 bits wide. This is particularly … Read more