arXiv subjects

Katsuhisa Ozaki

Publications and source records attributed to Katsuhisa Ozaki.

15 recordsLinked to original sources

arXiv2026

Ozaki Scheme II: A GEMM-oriented emulation of floating-point matrix multiplication using an integer modular technique

This paper addresses emulation algorithms for matrix multiplication. General Matrix-Matrix Multiplication (GEMM), a fundamental operation in the Basic Linear Algebra Subprograms (BLAS), is typically optimized for specific hardware architectures. The Ozaki scheme is a well-established GEMM-based emulation method for matrix multiplication, wherein input matrices are decomposed into several low-precision components to ensure that the resulting matrix product is computed exactly through numerical operations. This study proposes a novel GEMM-based emulation method for matrix multiplication that leverages the Chinese Remainder Theorem. The proposed method inherits the computational efficiency of highly optimized GEMM routines and further enables control over the number of matrix multiplications, which can enhance computational accuracy. We present numerical experiments featuring INT8 Tensor Core operations on GPUs and FP64 arithmetic on CPUs as case studies. The results demonstrate that FP64 emulation using the proposed method achieves performance levels of up to 7.4 to 9.8 TFLOPS on the NVIDIA RTX 4090 and 56.6 to 80.2 TFLOPS on the NVIDIA GH200, exceeding the measured performance of native FP64 arithmetic. Furthermore, for FP64 computations on CPUs, the proposed method achieved up to a 2.3x speedup in emulating quadruple-precision arithmetic compared to the conventional Ozaki scheme.

cs.MS↗

arXiv2026

Double-Precision Matrix Multiplication Emulation via Ozaki-II Scheme with FP8 Quantization

In this paper, we propose a method for emulating double-precision general matrix--matrix multiplication (DGEMM), a fundamental and performance-critical kernel in many high-performance computing applications. Ozaki-I and Ozaki-II are established DGEMM emulation schemes via low-precision matrix multiply-accumulate (MMA) units. For the Ozaki-I scheme, INT8-, FP8-, and FP16-based implementations have been proposed, all of which can be realized based on the same underlying algorithmic structure. In contrast, although INT8-based implementations of the Ozaki-II scheme have been reported, the original algorithm cannot be directly adapted to exploit FP8 MMA units. In several recent architectures, such as NVIDIA Blackwell Ultra and NVIDIA Rubin, INT8 performance has been reduced, making reliance on INT8 alone insufficient. Therefore, we introduce a novel technique to demonstrate DGEMM emulation based on the Ozaki-II scheme that operates on FP8 MMA units. Compared to the FP8-based Ozaki-I scheme, our method significantly reduces the computational cost and enables efficient FP64 emulation.

cs.DC↗

arXiv2026

Iterative Refinement for a Subset of Eigenvectors of Symmetric Matrices via Matrix Multiplications

We develop an iterative refinement method that improves the accuracy of a user-chosen subset of $k$ eigenvectors ($k\ll n$) of an $n\times n$ real symmetric matrix. Using an orthogonal matrix represented in compact WY form, the method expresses the eigenvector error through a correction matrix that can be approximated efficiently from Rayleigh quotients and residuals. Unlike refinement methods for a single eigenpair or for a full eigenbasis, the proposed method refines only the selected $k$ eigenvectors using $\mathcal{O}(nk)$ additional storage, and its dominant work can be organized as matrix--matrix multiplications. Under an eigenvalue separation condition, the refinement converges linearly; we also provide a conservative sufficient condition. Practical variants of the separation condition (e.g., via shifting) enable targeting other extremal parts of the spectrum. For tightly clustered eigenvalues, we discuss limitations and show that preprocessing can restore convergence in a representative sparse example. Numerical experiments on dense test matrices and sparse matrices from the SuiteSparse Matrix Collection illustrate attainable accuracy and problem-dependent convergence.

math.NA↗

arXiv2026

Forward Error-Oriented Iterative Refinement for Eigenvectors of a Real Symmetric Matrix

In this paper, we discuss numerical methods for the eigenvalue decomposition of real symmetric matrices. While many existing methods can compute approximate eigenpairs with sufficiently small backward errors, the magnitude of the resulting forward errors is often unknown. Consequently, when high-precision numerical solutions are required, the computational cost tends to increase significantly because backward errors must be reduced to an excessive degree. To address this issue, we propose an efficient approximation algorithm that aims to achieve a prescribed forward error, together with a high-accuracy numerical algorithm based on the Ozaki scheme -- an emulation technique for matrix multiplication -- adapted to this problem. Since the proposed method is not primarily focused on reducing backward errors, the computational cost can be significantly reduced. Finally, we present numerical experiments to evaluate the efficiency of the proposed method.

math.NA↗

arXiv2026

Error Analysis of Matrix Multiplication Emulation Using Ozaki-II Scheme

The Ozaki-II scheme is an emulation method that leverages the Chinese Remainder Theorem to compute high-precision matrix multiplication via a sequence of low-precision matrix multiplications. In this scheme, the attainable numerical accuracy improves as the number of low-precision matrix multiplications increases. Previous numerical studies have shown that single- and double-precision matrix multiplication using the Ozaki-II scheme achieves higher throughput than that of standard BLAS routines on modern AI hardware equipped with fast INT8 matrix multiply-accumulate units with INT8 inputs and INT32 accumulation. However, the accuracy of the Ozaki-II scheme can degrade when the exponent distribution of the input matrices is wide, in which case a large number of low-precision matrix multiplications is required to obtain high-precision results. In this paper, we present a rigorous deterministic error analysis of the Ozaki-II scheme. The proposed analysis not only clarifies the accuracy behavior of the method but also enables the estimation of the number of low-precision matrix multiplications required to achieve a desired level of numerical accuracy.

math.NA↗

arXiv2025

Emulation of Complex Matrix Multiplication based on the Chinese Remainder Theorem

Modern computing architectures feature low-precision matrix multiplication units that achieve substantially higher throughput than their high-precision counterparts. Motivated by this architectural trend, the emulation of high-precision matrix multiplication using low-precision hardware has attracted significant interest in the high-performance computing community. Ozaki, Uchino, and Imamura proposed the Ozaki-II scheme as a general framework for emulating matrix multiplication. Building on this framework, Uchino, Ozaki, and Imamura developed high-performance and power-efficient techniques for emulating single- and double-precision real matrix multiplication on INT8 matrix engines. Extending this line of research, the present study proposes high-performance emulation methods for single- and double-precision complex matrix multiplication on INT8 matrix engines, based on the Ozaki-II scheme. On an NVIDIA B200 GPU, the proposed methods achieve 4.4--6.5x and 4.0--5.6x speedups over the native single- and double-precision complex matrix multiplication routines from cuBLAS, respectively, for sufficiently large problem sizes. When lower accuracy than that of the standard routines is acceptable, the proposed methods can operate at even higher speed. Conversely, with only a modest increase in computation time, they can deliver higher accuracy than that of the standard routines. These properties suggest that the proposed approach has the potential to serve as a default algorithm across a wide range of applications.

cs.DC↗

arXiv2025

Sparse Iterative Solvers Using High-Precision Arithmetic with Quasi Multi-Word Algorithms

To obtain accurate results in numerical computation, high-precision arithmetic is a straightforward approach. However, most processors lack hardware support for floating-point formats beyond double precision (FP64). Double-word arithmetic (Dekker 1971) extends precision by using standard floating-point operations to represent numbers with twice the mantissa length. Building on this concept, various multi-word arithmetic methods have been proposed to further increase precision by combining additional words. Simplified variants, known as quasi algorithms, have also been introduced, which trade a certain loss of accuracy for reduced computational cost. In this study, we investigate the performance of quasi algorithms for double- and triple-word arithmetic in sparse iterative solvers based on the Conjugate Gradient method, and compare them with both non-quasi algorithms and standard FP64. We evaluate execution time on an x86 processor, the number of iterations to convergence, and solution accuracy. Although quasi algorithms require appropriate normalization to preserve accuracy - without it, convergence cannot be achieved - they can still reduce runtime when normalization is applied correctly, while maintaining accuracy comparable to full multi-word algorithms. In particular, quasi triple-word arithmetic can yield more accurate solutions without significantly increasing execution time relative to double-word arithmetic and its quasi variant. Furthermore, for certain problems, a reduction in iteration count contributes to additional speedup. Thus, quasi triple-word arithmetic can serve as a compelling alternative to conventional double-word arithmetic in sparse iterative solvers.

cs.MS↗

arXiv2025

High-Performance and Power-Efficient Emulation of Matrix Multiplication using INT8 Matrix Engines

Recent architectures integrate high-performance and power-efficient matrix engines. These engines demonstrate remarkable performance in low-precision matrix multiplication, which is crucial in deep learning. Several techniques have been proposed to emulate single- and double-precision general matrix-matrix multiplication (SGEMM and DGEMM, respectively) by leveraging such low-precision matrix engines. In this study, we present emulation methods that significantly outperforms conventional approaches. On a GH200 Grace Hopper Superchip, the proposed DGEMM emulation achieves a 1.4x speedup and a 43% improvement in power efficiency compared to native DGEMM for sufficiently large problems. The proposed SGEMM emulation achieves a 3.0x speedup and a 154% improvement in power efficiency compared to native SGEMM for sufficiently large problems. Furthermore, compared to conventional emulation methods, the proposed emulation achieves more than 2x higher performance and superior power efficiency.

cs.DC↗

arXiv2025

Verified error bounds for the singular values of structured matrices with applications to computer-assisted proofs for differential equations

This paper introduces two methods for verifying the singular values of the structured matrix denoted by $R^{-H}AR^{-1}$, where $R$ is a nonsingular matrix and $A$ is a general nonsingular square matrix. The first of the two methods uses the computed factors from a singular value decomposition (SVD) to verify all singular values; the second estimates a lower bound of the minimum singular value without performing the SVD. The proposed approach for verifying all singular values efficiently computes tight error bounds. The method for estimating a lower bound of the minimum singular value is particularly effective for sparse matrices. These methods have proven to be efficient in verifying solutions to differential equation problems, that were previously challenging due to the extensive computational time and memory requirements.

math.NA↗

arXiv2024

Reducing Numerical Precision Requirements in Quantum Chemistry Calculations

The abundant demand for deep learning compute resources has created a renaissance in low precision hardware. Going forward, it will be essential for simulation software to run on this new generation of machines without sacrificing scientific fidelity. In this paper, we examine the precision requirements of a representative kernel from quantum chemistry calculations: calculation of the single particle density matrix from a given mean field Hamiltonian (i.e. Hartree-Fock or Density Functional Theory) represented in an LCAO basis. We find that double precision affords an unnecessarily high level of precision, leading to optimization opportunities. We show how an approximation built from an error-free matrix multiplication transformation can be used to potentially accelerate this kernel on future hardware. Our results provide a road map for adapting quantum chemistry software for the next generation of High Performance Computing platforms.

physics.chem-ph↗

arXiv2024

Performance Enhancement of the Ozaki Scheme on Integer Matrix Multiplication Unit

This study was aimed at simultaneously achieving sufficient accuracy and high performance for general matrix multiplications. Recent architectures, such as NVIDIA GPUs, feature high-performance units designed for low-precision matrix multiplications in machine learning models, and next-generation architectures are expected to follow the same design principle. The key to achieving superior performance is to fully leverage such architectures. The Ozaki scheme, a highly accurate matrix multiplication algorithm using error-free transformations, enables higher-precision matrix multiplication to be performed through multiple lower-precision matrix multiplications and higher-precision matrix additions. Ootomo et al. implemented the Ozaki scheme on high-performance matrix multiplication units with the aim of achieving both sufficient accuracy and high performance. This paper proposes alternative approaches to improving performance by reducing the numbers of lower-precision matrix multiplications and higher-precision matrix additions. Numerical experiments demonstrate the accuracy of the results and conduct performance benchmarks of the proposed approaches. These approaches are expected to yield more efficient results in next-generation architectures.

cs.DC↗

arXiv2024

Method for Verifying Solutions of Sparse Linear Systems with General Coefficients

This paper proposes a verification method for sparse linear systems $Ax=b$ with general and nonsingular coefficients. A verification method produces the error bound for a given approximate solution. Conventional methods use one of two approaches. One approach is to verify the computed solution of the normal equation $A^TAx=A^Tb$ by exploiting symmetric and positive definiteness; however, the condition number of $A^TA$ is the square of that for $A$. The other approach uses an approximate inverse matrix of the coefficient; however, the approximate inverse may be dense even if $A$ is sparse. Here, we propose a method for the verification of solutions of sparse linear systems based on $LDL^T$ decomposition. The proposed method can reduce the fill-in and is applicable to many problems. Moreover, an efficient iterative refinement method is proposed for obtaining accurate solutions.

math.NA↗

arXiv2024

DGEMM on Integer Matrix Multiplication Unit

Deep learning hardware achieves high throughput and low power consumption by reducing computing precision and specializing in matrix multiplication. For machine learning inference, fixed-point value computation is commonplace, where the input and output values and the model parameters are quantized. Thus, many processors are now equipped with fast integer matrix multiplication units (IMMU). It is of significant interest to find a way to harness these IMMUs to improve the performance of HPC applications while maintaining accuracy. We focus on the Ozaki scheme, which computes a high-precision matrix multiplication by using lower-precision computing units, and show the advantages and disadvantages of using IMMU. The experiment using integer Tensor Cores shows that we can compute double-precision matrix multiplication faster than cuBLAS and an existing Ozaki scheme implementation on FP16 Tensor Cores on NVIDIA consumer GPUs. Furthermore, we demonstrate accelerating a quantum circuit simulation by up to 4.33 while maintaining the FP64 accuracy.

cs.DC↗

arXiv2024

Polygonal Sequence-driven Triangulation Validator: An Incremental Approach to 2D Triangulation Verification

Two-dimensional Delaunay triangulation is a fundamental aspect of computational geometry. This paper presents a novel algorithm that is specifically designed to ensure the correctness of 2D Delaunay triangulation, namely the Polygonal Sequence-driven Triangulation Validator (PSTV). Our research highlights the paramount importance of proper triangulation and the often overlooked, yet profound, impact of rounding errors in numerical computations on the precision of triangulation. The primary objective of the PSTV algorithm is to identify these computational errors and ensure the accuracy of the triangulation output. In addition to validating the correctness of triangulation, this study underscores the significance of the Delaunay property for the quality of finite element methods. Effective strategies are proposed to verify this property for a triangulation and correct it when necessary. While acknowledging the difficulty of rectifying complex triangulation errors such as overlapping triangles, these strategies provide valuable insights on identifying the locations of these errors and remedying them. The unique feature of the PSTV algorithm lies in its adoption of floating-point filters in place of interval arithmetic, striking an effective balance between computational efficiency and precision. This research sets a vital precedent for error reduction and precision enhancement in computational geometry.

cs.CG↗

arXiv2020

An a posteriori verification method for generalized real-symmetric eigenvalue problems in large-scale electronic state calculations

An a posteriori verification method is proposed for the generalized real-symmetric eigenvalue problem and is applied to densely clustered eigenvalue problems in large-scale electronic state calculations. The proposed method is realized by a two-stage process in which the approximate solution is computed by existing numerical libraries and is then verified in a moderate computational time. The procedure returns intervals containing one exact eigenvalue in each interval. Test calculations were carried out for organic device materials, and the verification method confirms that all exact eigenvalues are well separated in the obtained intervals. This verification method will be integrated into EigenKernel (https://github.com/eigenkernel/), which is middleware for various parallel solvers for the generalized eigenvalue problem. Such an a posteriori verification method will be important in future computational science.

physics.comp-ph↗