Multithreaded implicitly dealiased convolutions
نویسندگان
چکیده
Implicit dealiasing is a method for computing in-place linear convolutions via fast Fourier transforms that decouples work memory from input data. It offers easier memory management and, for long one-dimensional input sequences, greater efficiency than conventional zero-padding. Furthermore, for convolutions of multidimensional data, the segregation of data and work buffers can be exploited to reduce memory usage and execution time. This is accomplished by processing and discarding data as it is generated, allowing work memory to be reused, for greater data locality and performance. A multithreaded implementation of implicit dealiasing that accepts an arbitrary number of input and output vectors and a general multiplication operator is presented, along with an improved one-dimensional Hermitian convolution that avoids the loop dependency inherent in previous work. An alternate data format that can accommodate a Nyquist mode and enhance cache efficiency is also proposed.
منابع مشابه
Implicitly Dealiased Convolutions: Example Applications and Performance Comparison
Implicitly dealiasing is a recently-developed technique which improves upon conventional zero padding to compute linear convolutions via fast Fourier transforms. For onedimensional inputs, the memory requirements and performance are similar to conventional zero-padded convolutions, but implicitly dealiased convolutions are faster and require less memory when the data is multi-dimensional. We sh...
متن کاملMultithreaded Implicitly Dealiased Pseudospectral Convolutions
Convolutions are used in data and image analysis and form the crux of the pseudospectral method for direct numerical simulations of fluids. Their calculation is a computationally expensive task that is facilitated by the use of FFTs at the expense of increased memory, which is required for the removal of aliased terms. Here, we present a multithreaded version of the method of implicit dealiasin...
متن کاملEfficient Dealiased Convolutions without Padding
Algorithms are developed for calculating dealiased linear convolution sums without the expense of conventional zero-padding or phase-shift techniques. For one-dimensional in-place convolutions, the memory requirements are identical with the zero-padding technique, with the important distinction that the additional work memory need not be contiguous with the input data. This decoupling of data a...
متن کاملDealiased Convolutions for Pseudospectral Simulations
Efficient algorithms have recently been developed for calculating dealiased linear convolution sums without the expense of conventional zero-padding or phase-shift techniques. For one-dimensional in-place convolutions, the memory requirements are identical with the zero-padding technique, with the important distinction that the additional work memory need not be contiguous with the input data. ...
متن کاملCasimir Cascades in Two-Dimensional Turbulence
In addition to conserving energy and enstrophy, the nonlinear terms of the twodimensional incompressible Navier–Stokes equation are well known to conserve the global integral of any continuously differentiable function of the scalar vorticity field. However, the phenomenological role of these additional inviscid invariants, including the issue as to whether they cascade to large or small scales...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید
ثبت ناماگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید
ورودعنوان ژورنال:
- J. Comput. Physics
دوره 356 شماره
صفحات -
تاریخ انتشار 2018