How Flexible is Your Computing System?

FOS: Computer and information sciences metric Hardware Architecture (cs.AR) 0202 electrical engineering, electronic engineering, information engineering SDG 7 - Affordable and Clean Energy 02 engineering and technology Flexibility versatility Computer Science - Hardware Architecture SDG 7 – Betaalbare en schone energie
DOI: 10.1145/3524861 Publication Date: 2022-03-30T11:35:16Z
ABSTRACT
In literature, computer architectures are frequently claimed to be highly flexible , typically implying the existence of trade-offs between flexibility and performance or energy efficiency. Processor flexibility, however, is not very sharply defined, consequently these claims cannot validated, nor can such hypothetical relations fully understood exploited in design computing systems. This paper an attempt introduce scientific rigour notion A survey conducted provide overview references both architecture domain, as well related fields. classification introduced categorize different views on which ultimately form foundation for a qualitative definition flexibility. Departing from generic quantifiable metric proposed, enabling valid quantitative comparison various architectures. To validate proposed method, evaluate relation general measured 25 systems, including CPUs, GPUs, DSPs, FPGAs, 40 ASIPs taken literature. The obtained results insights into some speculative properties efficiency area Overall shows commensurate with generally accepted notions collected survey, although surprising discrepancies also observed. placed context state art compute extensive reflection provides only complete field, but discusses possible alternative approaches open issues. Note that this work does aim final answer rather framework initiate broader discussion society defining, understanding, taking advantage
SUPPLEMENTAL MATERIAL
Coming soon ....
REFERENCES (56)
CITATIONS (4)
EXTERNAL LINKS
PlumX Metrics
RECOMMENDATIONS
FAIR ASSESSMENT
Coming soon ....
JUPYTER LAB
Coming soon ....