Truncated SIMD Multiplier Architecture for Approximate Computing in Low-Power Programmable Processors
Use este enlace para citar
http://hdl.handle.net/2183/34895
A non ser que se indique outra cousa, a licenza do ítem descríbese como Atribución-NoComercial-SinDerivadas 4.0 Internacional ( CC-BY-NC-ND)
Coleccións
- GI-GAC - Artigos [193]
Metadatos
Mostrar o rexistro completo do ítemTítulo
Truncated SIMD Multiplier Architecture for Approximate Computing in Low-Power Programmable ProcessorsData
2019Cita bibliográfica
R. R. Osorio and G. Rodríguez, "Truncated SIMD Multiplier Architecture for Approximate Computing in Low-Power Programmable Processors," in IEEE Access, vol. 7, pp. 56353-56366, 2019, doi: 10.1109/ACCESS.2019.2913743.
Resumo
[Abstract]: Approximate computing has been exploited for many years in application-specific architectures. Recently, it has also been proposed for low-power programmable processors. However, this poses some challenges as, in a microprocessor, the energy consumed by fetching and decoding an instruction may be significantly higher than that of the execution itself. Therefore, approximate computing would be advisable only for those instructions, in which the execution stage is significantly expensive in terms of energy consumption. In this paper, we present new architectures for truncated SIMD multipliers able to calculate signed and unsigned products from 8 × 8 to 64× 64 bits. Next, we analyze the precision loss incurred by truncation for all product sizes. We implement accurate and truncated architectures for both scalar and SIMD products and find that truncation allows area savings of up to 27%. The proposed design is experimentally evaluated in different scenarios, showing potential energy savings ranging from 29% to 42%. Finally, this paper analyzes the overall convenience of introducing truncated SIMD architectures with respect to accurate SIMD and scalar architectures.
Palabras chave
Digital arithmetic
Fixed-point arithmetic
Approximate computing
Approximate multiplier
Low power
Fixed-point arithmetic
Approximate computing
Approximate multiplier
Low power
Versión do editor
Dereitos
Atribución-NoComercial-SinDerivadas 4.0 Internacional ( CC-BY-NC-ND)
ISSN
2169-3536