Les unités de calcul à virgule flottante (FP) prennent en charge un sous-ensemble de formats définis par la norme IEEE 754, ainsi que quelques formats qui leur sont spécifiques (le format de 80 bits sur de l’architecture x86, et le format 128 bit propriétaire des PowerPC). De fait, les langages et les compilateurs optimisants utilisés en calcul intensif sont limités par les formats supportés sur les machines cibles. Cependant, la recherche de l’efficacité et de la stabilité des applications a conduit les numériciens à explorer d’autres tailles pour les exposants et les parties fractionnaires afin de trouver un bon équilibre entre la précision des résultats, le temps d’exécution et l’énergie consommée. C’est le cas pour les calculs numériques qui ...
High-level synthesis (HLS) tools offer increased productivity regarding FPGA programming. However, d...
Numerical accuracy and execution time of programs using the floating-point arithmetic are major chal...
Media processing applications typically involve numerical blocks that exhibit regular floating-point...
Floating-Point (FP) units in processors are generally limited to supporting a subset of formats defin...
Most of the Floating-Point (FP) hardware units support the formats and the operations specified in t...
International audienceFloating-Point (FP) units in processors are generally limited to supporting a ...
International audienceOptimizing compilers for high performance computing only support IEEE 754 floa...
Le Calcul Haute Performance (HPC) est un écosystème dynamique où architectures et codes de calcul sc...
Full-precision Floating-Point Units (FPUs) can be a source of extensive hardware overhead in general...
Since the apparition of the first computer, floating point arithmetic have drastically changed. The ...
International audienceFloating-point hardware support has more or less been settled 35 years ago by ...
International audienceThe popularity and community-driven development model of RISCV have opened man...
High-level synthesis (HLS) tools offer increased productivity regarding FPGA programming. However, d...
Numerical accuracy and execution time of programs using the floating-point arithmetic are major chal...
Media processing applications typically involve numerical blocks that exhibit regular floating-point...
Floating-Point (FP) units in processors are generally limited to supporting a subset of formats defin...
Most of the Floating-Point (FP) hardware units support the formats and the operations specified in t...
International audienceFloating-Point (FP) units in processors are generally limited to supporting a ...
International audienceOptimizing compilers for high performance computing only support IEEE 754 floa...
Le Calcul Haute Performance (HPC) est un écosystème dynamique où architectures et codes de calcul sc...
Full-precision Floating-Point Units (FPUs) can be a source of extensive hardware overhead in general...
Since the apparition of the first computer, floating point arithmetic have drastically changed. The ...
International audienceFloating-point hardware support has more or less been settled 35 years ago by ...
International audienceThe popularity and community-driven development model of RISCV have opened man...
High-level synthesis (HLS) tools offer increased productivity regarding FPGA programming. However, d...
Numerical accuracy and execution time of programs using the floating-point arithmetic are major chal...
Media processing applications typically involve numerical blocks that exhibit regular floating-point...