Search Results for "thread"

...It incorporates strategies for hierarchical decomposition and data movement similar to those used to implement cuBLAS and cuDNN. CUTLASS decomposes these "moving parts" into reusable, modular software components abstracted by C++ template classes. These thread-wide, warp-wide, block-wide, and device-wide primitives can be specialized and tuned via custom tiling sizes, data types, and other algorithmic policy. The resulting flexibility simplifies their use as building blocks within custom kernels and applications. To support a wide variety of applications, CUTLASS provides extensive support for mixed-precision computations, providing specialized data-movement and multiply-accumulate abstractions for half-precision floating point (FP16), BFloat16 (BF16), Tensor Float 32 (TF32), etc.

...Features high-performance Numba-accelerated search and a signature 'film-grain' aesthetic upon reconstruction. ZIP also include 16 MB file with 16,7 mil numbers of PI Benchmark(Single-Thread): Hardware & Environment Apple Silicon: Apple M2 (Mac mini/MacBook) x86_64 Platform: Intel Core Ultra 5 225F (Arrow Lake, 10 Cores) OS 1: Fedora 43 (GNOME) OS 2: Windows 11 Pro (23H2/24H2) Software: Python 3.14.3 + Numba JIT (latest) Results (Lower is better) Platform / OS CPU Time (Seconds) macOS (Native) Apple M2 52.151311 s (in default setup) Fedora Linux Intel Core Ultra 5 225F 58.536457 s (in default Power Management: Balanced) Windows 11 Intel Core Ultra 5 225F 59.681427 s (important! ...

A Java package that provides routines for various statistical distributions. Based on R version 2.14.1 (continuously updated; current as of R v3.3.0). The major difference is that JDistlib is thread safe. The library contains the density (pdf), cumulative (cdf), quantile, and random number generator (RNG) routines of the following distributions: Ansari, Beta, Binomial, Cauchy, Chi square, Exponential, Fisher's F, Gamma, Geometric, Hypergeometric, Kendall, Logistic, Log normal, Negative binomial, Noncentral beta, Noncentral chi square, Noncentral f, Noncentral t, Normal, Poisson, Sign Rank, Spearman, Student's T, Tukey, Uniform, Weibull, Wilcoxon, and many more. ...

The first parallel and distributed library for pairwise correlation/dependence computation on Intel Xeon Phi clusters. This library is written in C++ template classes and achieves high speed by exploring the SIMD-instruction-level and thread-level parallelism within Xeon Phis as well as accelerator-level parallelism among multiple Xeon Phis. To facilitate balanced workload distribution, we have proposed a general framework for symmetric all-pairs computation by building provable bijective functions between job identifier and coordinate space for the first time.

Jafaran (Java Fast Random) provides fast, and for some more random, implementations of java.util.Random, with additional nextXXX() methods, and methods to retrieve and restore state. The names of implementations contain "Conc" (for concurrent) if they are thread-safe and non-blocking, or "Seq" (for sequential) if they are not thread-safe. Also provides an implementation of Ziggurat algorithm (based on J. A. Doornik paper, 2005), used by nextGaussian() methods of the provided implementations. Requires Java 5 or later. Also available on github since 2015/12/13: https://github.com/jeffhain/jafaran Principal classes: - Implementations using Mersenne-Twister algorithm (good pseudo-randomness): MTSyncRNG MTSeqRNG - Implementations using Marsaglia Xor-Shift (fast): MXSIntSeqRNG (32 bits) MXSLongSeqRNG (64 bits) (nextLong() faster, larger period) - Ziggurat: Random-based implementation of Ziggurat algorithm.

Trad4 is a fully concurrent, thread safe, graph-based programming language that scales linearly on multiple cores. It is initially intended for deployment in the financial industry to model real-time risk.

libfgen is a library that implements an efficient and customizable genetic algorithm (GA). It also provides particle swarm optimization (PSO) functionality and an interface for real-valued function minimization or model fitting. It is written in C, but can also be compiled with a C++ compiler. Both Linux and Windows are supported.

Low-latency high speed hybrid framework-library for utilizing the IBM Cell processor SPE units (PS3, QS20) to carry out computational tasks. Provides some DSP tasks such as FFT and complex MAC. Thread-safe and easily extensible with new arithmetics tasks

Platform independend, multithreaded fractal generator. Fine control over thread/CPU usage; unlimited precision; smooth/random/function coloring; supports color map files (over 1600 available); many image file formats. For any Java 1.5+ (32/64 Bit) VM.

Glucas is a client to make primality test for Mersenne numbers. It is related to GIMPS effort to search the bigest primes ever found. In addition, it can be used as an excellent performance test. Posix thread, OpenMP and SunMP compatible