Cache-oblivious sparse matrix–vector multiplication by using sparse matrix partitioning methods
AN Yzelman, RH Bisseling - SIAM Journal on Scientific Computing, 2009 - SIAM
In this article, we introduce a cache-oblivious method for sparse matrix–vector multiplication.
Our method attempts to permute the rows and columns of the input matrix using a recursive …
Our method attempts to permute the rows and columns of the input matrix using a recursive …
Japanese autotuning research: Autotuning languages and FFT
T Katagiri, D Takahashi - Proceedings of the IEEE, 2018 - ieeexplore.ieee.org
This paper introduces current research on automatic performance tuning, specifically in the
Japanese community, from two aspects. First, we discuss autotuning (AT) research from the …
Japanese community, from two aspects. First, we discuss autotuning (AT) research from the …
Auto-tuning on NUMA and many-core environments with an FDM code
T Katagiri, S Ohshima… - 2017 IEEE International …, 2017 - ieeexplore.ieee.org
In this paper, we focus on auto-tuning (AT) performance on nonuniform memory access
(NUMA) and many-core architectures. Code from the finite difference method (FDM) is …
(NUMA) and many-core architectures. Code from the finite difference method (FDM) is …
TuningGenie: auto-tuning framework based on rewriting rules
This paper presents results on development of the auto-tuning framework named
TuningGenie aimed at automating adjustment of parallel tasks to target platform. The …
TuningGenie aimed at automating adjustment of parallel tasks to target platform. The …
Directive-based auto-tuning for the finite difference method on the Xeon Phi
T Katagiri, S Ohshima… - 2015 IEEE International …, 2015 - ieeexplore.ieee.org
In this paper, we present a directive-based auto-tuning (AT) framework, called ppOpen-AT,
and demonstrate its effect using simulation code based on the Finite Difference Method …
and demonstrate its effect using simulation code based on the Finite Difference Method …
Voltage-scaling scheduling for periodic real-time tasks in reward maximization
This paper is interested in reward maximization of periodic real-time tasks under a given
energy constraint, where the reward received depends on how much computation a task …
energy constraint, where the reward received depends on how much computation a task …
Auto-tuning of computation kernels from an FDM Code with ppOpen-AT
T Katagiri, S Ohshima… - 2014 IEEE 8th …, 2014 - ieeexplore.ieee.org
In this paper, we propose an Auto-tuning (AT) function with an AT language for a dedicated
numerical library with respect to supercomputers in operation. The AT function is based on …
numerical library with respect to supercomputers in operation. The AT function is based on …
Implementation of D‐Spline‐Based Incremental Performance Parameter Estimation Method with ppOpen‐AT
T Tanaka, R Otsuka, A Fujii, T Katagiri… - Scientific …, 2014 - Wiley Online Library
In automatic performance tuning (AT), a primary aim is to optimize performance parameters
that are suitable for certain computational environments in ordinary mathematical libraries …
that are suitable for certain computational environments in ordinary mathematical libraries …
Early experiences for adaptation of auto-tuning by ppOpen-AT to an explicit method
T Katagiri, S Ito, S Ohshima - 2013 IEEE 7th International …, 2013 - ieeexplore.ieee.org
We present a code optimization technique by adapting an auto-tuning (AT) function to an
explicit method with the static code generator FIBER. The AT function is evaluated with …
explicit method with the static code generator FIBER. The AT function is evaluated with …
Auto-tuning Mixed-precision Computation by Specifying Multiple Regions
X Ren, M Kawai, T Hoshino, T Katagiri… - … on Computing and …, 2023 - ieeexplore.ieee.org
Mixed-precision computation is a promising method for substantially increasing the speed of
numerical computations. However, using mixed-precision data is a double-edged sword …
numerical computations. However, using mixed-precision data is a double-edged sword …