Nanos6 is a runtime that implements the OmpSs-2 parallel programming model, developed by the System Tools and Advanced Runtimes (STAR) group at the Barcelona Supercomputing Center.

This meta-repository contains the releases of the OmpSs-2 programming model

The Accelerator Integration Tool (AIT) automatically integrates OmpSs@FPGA accelerators into FPGA designs using different vendor backends

The Task-Aware CUDA (TACUDA) provides interoperability support between task-based programming models and CUDA which enables the taskification of CUDA operations and kernels on NVIDIA accelerators

The Task-Aware SYCL (TASYCL) provides interoperability support between task-based programming models and SYCL which enables the taskification of SYCL operations and kernels on accelerators

The Task-Aware HIP (TAHIP) provides interoperability support between task-based programming models and HIP which enables the taskification of HIP operations and kernels on accelerators

Main set of benchmarks for OmpSs-2@Cluster

Library implementing a common interface to manage FPGA memory and streams

OmpSs-2@Clusters small tutorial code.

The Task-Aware AscendCL (TACL) provides interoperability support between task-based programming models and CUDA which enables the taskification of AscendCL operations and kernels on Huawei Ascend accelerators

N-body simulation is a simulation of a dynamical system of particles, usually under the influence of physical forces, such as gravity.

Meta-repository for OmpSs-2@FPGA releases

Bare implementation of static containers that can be used in OmpSs-2@Clusters.

This is a port of SPH-EXA application for OmpSs-2 task-based parallel programming model. SPH-EXA is an application contained in SpecHPC 2021 benchmark suite.

This application performs a cholesky decomposition/factorization over a square matrix. The matrix is distributed by blocks of contiguous memory.

Adapter to emit instrumentation events from accelerators

Runtime system for OmpSs-2@Cluster programming model

Library implementing a common interface to manage FPGA tasks

Support module for creating tasks from accelerators

This application performs the multiplication of two square matrices. The matrices are allocated by blocks of contiguous memory.