arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2406.05398cs.AR
收藏

使用软件定义数据流架构评估 Posit 在谱分析中的表现

Evaluation of Posits for Spectral Analysis Using a Software-Defined Dataflow Architecture

Sameer Deshmukh, Daniel Khankin, William Killian, John Gustafson, Elad Raz

AI总结:

本文提出用软件定义数据流架构公平评估 posit 与 IEEE 754 在谱分析中的性能,证明 posit 精度更高且在该架构上仅慢 1.8 倍,并给出新的性能下界。

摘要:

谱分析在结构损伤检测和深度学习中发挥着重要作用。浮点格式的选择对谱分析的精度和性能起着关键作用。IEEE Std 754™ 浮点格式(简称 IEEE 754)得到了大多数主要硬件厂商对“普通”浮点数的支持,但它存在若干局限性。此前的工作已尝试从精度和性能方面评估 posit 格式。在多种应用中,posit 的精度已被证明优于 IEEE 754。例如,我们对快速傅里叶变换的分析表明,使用 32 位 posit 相比 32 位 IEEE 754 格式可获得 2 倍更好的精度。对于谱分析,32 位 posit 明显比 32 位 IEEE 754 浮点数更精确。尽管 posit 已显示出比 IEEE 754 更好的精度,但迄今为止仍缺乏在真实硬件实现上对 posit 与 IEEE 754 格式进行公平评估的工作。在 x86 CPU 上对 posit 格式进行软件仿真时,对于 2^28 个点的快速傅里叶变换(FFT),其速度比原生 IEEE 754 普通浮点硬件慢约 69.3 倍。我们提出使用软件定义数据流架构来评估 posit 在谱分析中的性能和精度。我们的数据流架构使用可重构逻辑单元,仅用整数运算来表达算法。我们的架构没有 FPU,并且在硬件内使用相同的整数运算来表达 IEEE 754 和 posit 算术。在我们的数据流架构上,对于 2^28≈2.68 亿个点的快速傅里叶变换(FFT),posit 格式仅比 IEEE 754 慢 1.8 倍。借助这一实现,我们凭经验为 posit 相对于 IEEE 754 格式的性能提出了一个新的下界。

英文摘要:

Spectral analysis plays an important role in detection of damage in structures and deep learning. The choice of a floating-point format plays a crucial role in determining the accuracy and performance of spectral analysis. The IEEE Std 754\textsuperscript{TM} floating-point format (IEEE~754 for short) is supported by most major hardware vendors for ``normal'' floats. However, it has several limitations. Previous work has attempted to evaluate posit format with respect to accuracy and performance. The accuracy of the posit has been established over IEEE~754 for a variety of applications. For example, our analysis of the Fast Fourier Transform shows 2x better accuracy when using a 32-bit posit vs. a 32-bit IEEE754 format. For spectral analysis, 32-bit posits are substantially more accurate than 32-bit IEEE~754 floats. Although posit has shown better accuracy than IEEE~754, a fair evaluation of posit with IEEE~754 format using a real hardware implementation has been lacking so far. A software simulation of posit format on an x86 CPU is about $\mathbf{69.3\times}$ slower than native IEEE~754 hardware for normal floats for a Fast Fourier Transform (FFT) of $\mathbf{2^{28}}$ points. We propose the use of a software-defined dataflow architecture to evaluate performance and accuracy of posits in spectral analysis. Our dataflow architecture uses reconfigurable logical elements that express algorithms using only integer operations. Our architecture does not have an FPU, and we express both IEEE~754 and posit arithmetic using the same integer operations within the hardware. On our dataflow architecture, the posit format is only $\mathbf{1.8\times}$ slower than IEEE~754 for a Fast Fourier Transform (FFT) of $\mathbf{2^{28}\approx 268}$ million points. With this implementation, we empirically propose a new lower bound for the performance of posit compared to IEEE~754 format.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑