SOFTWARE & HARDWARE

Licensable technology created by BSC

Showing 81 - 100 results of 159

Provides several features to integrate other programming models and low-level APIs with the data-flow execution model of OpenMP tasks.

lumi-g-test project is a set of C++ micro-benchmarks that helps to test several important harware and system software characteristics of the GPU partition of LUMI supercomputer (https://www.lumi-supercomputer.eu). The metrics that are estimated by micro-benchmarks are mostly related of data transfer bandwidth aspects. The benchmarks cover cross-node and intra-node data tranfsers using MPI library, host memory to GPU memory transfer bandwidth characteristics, GPU memory internal bandwidth.

Tool for the estimation of probabilistic WCET based on execution time measurements (in the form of an R script). 

Details of the method available in: https://doi.org/10.1145/3065924

Coyote is an execution-driven simulator based on the open source RISC-V standard, built upon two preexisting simulation tools: Spike and Sparta.

Solderpad Hardware License (Version 2.1)

A flexible, configurable and portable set of FPGA- IPs that enables any emulated design to communicate with the host and other FPGAs out of the box.

The FPGA shell is meant to be a static perimeter architecture that guarantees that the inside accelerator package can be interchangeable for any other package when meeting a defined I/O interface between the shell and the accelerator package.

 

Modification of the linux kernel that allows to deduplicate memory pages (page cache level). This increases the memory available for cache I/O while being transparent for the user and applications.

Mercurium is a source-to-source compilation infrastructure aimed at fast prototyping. Current supported languages are C99, C++11 and Fortran 95. Mercurium is mainly used along with the Nanos++ runtime to implement projects for OmpSs and OpenMP but since it is quite extensible it has been used in other projects including (but not limiting to) Cell Superscalar, ACOTES, software transactional memory, vectorization and correctness.

Tools to Merge Hardware Event Monitors (HEMs) Coming from Separate Subexperiments into One Single Dataframe.

Implementation of two tools to merge Hardware Event Monitors (HEMs) from different subexperiments. Hardware Reading and Merging (HRM), which uses order statistics to merge; and MUlti-Correlation HEM (MUCH) which merges using a multivariate normal distribution. 

The Multiscale Online Nonhydrostatic AtmospheRe CHemistry model (MONARCH) is a chemical weather prediction system, formerly known as NMMB/BSC-CTM, that can be run either globally or regionally. MONARCH, developed at the Barcelona Supercomputing Center, is based on the online coupling of the meteorological Nonhydrostatic Multiscale Model on the B-grid (NMMB) with a full aerosol-chemistry module. Under continuous development. Used as operational model in the WMO Barcelona Dust Forecast Center and the Copernicus Regional Production service.

Multi-cores in real-time systems: opportunities and challenges
Multi-core processors are becoming the baseline computing solution in critical embedded systems. While multi-cores allow high software integration levels, hence reducing hardware procurement and SWaP (Space, Weight and Power) costs, their use challenge current practices in timing analysis.

Nanos++ is a runtime designed to serve as runtime support in parallel environments. It is mainly used to support  OmpSs, a extension to OpenMP developed at BSC. It also has modules to support  OpenMP 3.1.

GPL License (Version 3.0)

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.
NetCleave is a bioinformatic tool to aid in protein cleavage predictions by the Proteasome. More in detail, it is an open-source and retrainable algorithm for the prediction of the C-terminal antigen processing for both MHC-I and MHC-II pathways. NetCleave architecture consists of a neural network trained on 46 different physicochemical descriptors of the amino acids forming the cleavage site. 

GPL License (Version 3.0)

NODES -- short for nOS-V-based Dependency System -- is a runtime library designed to work on top of the nOS-V runtime while providing most of the functionalities from its predecessor, Nanos6. NODES acts as a dependency system for nOS-V by synchronizing the data flow of codes annotated with OmpSs-2 directives and bridges the gap between nOS-V and user applications through these annotations. Simultaneously, nOS-V manages the life cycle of multiple generic tasks from various applications or libraries such as NODES and handles their interaction with the system.

GPL License (Version 3.0)

nOS-V is a lightweight tasking library. Users can create tasks and submit them for execution, and nOS-V will schedule tasks in the available resources. The novelty of nOS-V is that multiple independent applications can share a single nOSV “instance”, giving the library a system-wide view of tasks and resources and enabling what we call “co-execution”. Co-execution is a paradigm for two or more applications to share resources using a single scheduler, which permits sharing resources at a granularity optimal for task-based applications.
Set of NUMA-aware schedule policies for the Nanos++ runtime system. Includes DEP (simple NUMA-aware scheduler) and RIP (based on graph partitioning techniques).

LLVM support for OpenMP offloading to NVIDIA BlueField DPUs via the DOCA API.

OmpSs is an effort to integrate features from the StarSs programming model developed by BSC into a single programming model. Combination of Mercurium and Nanos6.

ONCOLINER is an integrated platform with benchmarking data and tools for the detailed assessment, improvement and quality-based harmonization of analysis pipelines across centers. It can not only improve the overall efficiency of somatic variant identification globally, but it will also enable interoperability and consistency within emerging multi-center and multi-hospital data environments, allowing the sharing and integration of cancer datasets and results.

Provides several features to integrate other programming models and low-level APIs with the data-flow execution model of OpenMP tasks.

Pages