Scalable Parallel Numerical CSP Solver

نویسندگان

  • Daisuke Ishii
  • Kazuki Yoshizoe
  • Toyotaro Suzumura
چکیده

We present a parallel solver for numerical constraint satisfaction problems (NCSPs) that can scale on a number of cores. Our proposed method runs worker solvers on the available cores and simultaneously the workers cooperate for the search space distribution and balancing. In the experiments, we attained up to 119-fold speedup using 256 cores of a parallel computer.

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

A Scalable Parallel Assembly for Irregular Meshes Based on a Block Distribution for a Parallel Block Direct Solver

This paper describes a distribution of elements for irregular nite element meshes as well as the associated parallel assembly algorithm , in the context of parallel solving of the resulting sparse linear system using a direct block solver. These algorithms are integrated in the software processing chain EMILIO being developped at LaBRI for structural mechanics applications. Some illustrative nu...

متن کامل

Zurich ̈ Technische Hochschule

We present PARDISO, a new scalable parallel sparse direct linear solver on shared memory multiprocessors. In this paper, we describe the parallel fac-torization algorithm which utilizes the supernode structure of the matrix to reduce the number of memory references with Level 3 BLAS. We also propose enhancements that signiicantly reduce the communication rate for pipelining parallelism. The res...

متن کامل

Scalable Arbitrary-Order Pseudo-Spectral Electromagnetic Solver

Numerical simulations have been critical in the recent rapid developments of advanced accelerator concepts. Among the various available numerical techniques, the ParticleIn-Cell (PIC) approach is the method of choice for self-consistent simulations from first principles, and although the fundamentals of the PIC method were established decades ago, improvements or variations still continue to ar...

متن کامل

On Techniques to Improve Robustness and Scalability of a Parallel Hybrid Linear Solver

A hybrid linear solver based on the Schur complement method has great potential to be a general purpose solver scalable on tens of thousands of processors. For this, it is imperative to exploit two levels of parallelism; namely, solving independent subdomains in parallel and using multiple processors per subdomain. This hierarchical parallelism can lead to a scalable implementation which mainta...

متن کامل

AI AA 93 - 31 79 Performance of a Characteristic - Based , 3 - D , Time - Domain Maxwell Equations Solver on a Massively Parallel Computer

A characteristic-based. windward numerical prccerlure for solving three-dimensional Maxwell equations iii the time domain has been successfully ported to the Ititpi Touchstone Delta ninlticoniputer. The riumerical results by concurrent computation duplicated the earlier simulations of an oscillating electric dipole on a vector processor and compared w~11 with the exact solutioti. 'I'lie paralle...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2014