In this paper, we present a GPU-accelerated hybrid system that solves ill-conditioned systems of linear equations exactly. Exactly means without rounding errors due to using integer arithmetics. First, we scale floating-point numbers up to integers, then we solve dozens of SLEs within different modular arithmetics and then we assemble sub-solutions back using the Chinese remainder theorem. This approach effectively bypasses current CPU floating-point limitations. The system is capable of solving Hilbert’s matrix without losing a single bit of precision, and with a significant speedup compared to existing CPU solvers
Due to non-associativity of floating-point operations and dynamic scheduling on parallel architectur...
Parallelism in today's computer architectures is ubiquitous whether it be in supercomputers, worksta...
Motivated by the demand in machine learning, modern computer hardware is increas- ingly supporting r...
In this paper, we present a GPU-accelerated hybrid system that solves ill-conditioned systems of lin...
In this "Habilitation à Diriger des Recherches" (HDR), we present our research in high-performance s...
The Reliable Computing journal has no more paper publication, only free, electronic publication.Inte...
We have advanced the application of algorithms within a method of basic matrices, which are equipped...
GPUs (Graphics Processing Units) have become one of the main co-processors that contributed to deskt...
GPUs (Graphics Processing Units) have become one of the main co-processors that contributed to deskt...
Since its standardization in 1985, floating-point arithmetic is commonly used toapproximate computat...
Since 1985, the IEEE 754 standard defines formats, rounding modes and basic operations for floating-...
We present several algorithms to compute the solution of a linear system of equations on a graphics ...
C-XSC is a powerful C++ class library which simplifies the development of selfverifying numerical so...
Abstract—Since 1985, the IEEE 754 standard defines for-mats, rounding modes and basic operations for...
International audienceThis paper studies the performance of different algorithms for solving a dense...
Due to non-associativity of floating-point operations and dynamic scheduling on parallel architectur...
Parallelism in today's computer architectures is ubiquitous whether it be in supercomputers, worksta...
Motivated by the demand in machine learning, modern computer hardware is increas- ingly supporting r...
In this paper, we present a GPU-accelerated hybrid system that solves ill-conditioned systems of lin...
In this "Habilitation à Diriger des Recherches" (HDR), we present our research in high-performance s...
The Reliable Computing journal has no more paper publication, only free, electronic publication.Inte...
We have advanced the application of algorithms within a method of basic matrices, which are equipped...
GPUs (Graphics Processing Units) have become one of the main co-processors that contributed to deskt...
GPUs (Graphics Processing Units) have become one of the main co-processors that contributed to deskt...
Since its standardization in 1985, floating-point arithmetic is commonly used toapproximate computat...
Since 1985, the IEEE 754 standard defines formats, rounding modes and basic operations for floating-...
We present several algorithms to compute the solution of a linear system of equations on a graphics ...
C-XSC is a powerful C++ class library which simplifies the development of selfverifying numerical so...
Abstract—Since 1985, the IEEE 754 standard defines for-mats, rounding modes and basic operations for...
International audienceThis paper studies the performance of different algorithms for solving a dense...
Due to non-associativity of floating-point operations and dynamic scheduling on parallel architectur...
Parallelism in today's computer architectures is ubiquitous whether it be in supercomputers, worksta...
Motivated by the demand in machine learning, modern computer hardware is increas- ingly supporting r...