SOFTWARE & HARDWARE
Licensable technology created by BSC
The technology in question is the High-Elective Resolution Modelling Emission System version 3 (HERMESv3), a Python-based and multi-scale modeling tool designed for processing and computing atmospheric emissions for air quality modeling. HERMESv3 consists of two modules: the global_regional module and the bottom_up module. The focus of the bottom_up module is on estimating anthropogenic emissions at high spatial (e.g., road link level) and temporal (hourly) resolutions. It employs advanced calculation methods, combining local activity, emission factors, and meteorological data. This technology facilitates the computation of emissions from various sources, including point sources, road transport, combustion activities, and agriculture, providing a versatile and adaptable solution for estimating air pollutants and greenhouse gases in European countries or regions with the required input data.
The High-Elective Resolution Modelling Emission System version 3 global regional module (HERMESv3_GR) is an open source, parallel and stand-alone multi-scale atmospheric emission processing system that computes gaseous and aerosol emissions for use in atmospheric chemistry models. HERMESv3_GR is coded in Python and calculates emissions from different sources, regions and pollutants on a user-specified global or regional grid.
Horus is an innovative multi-platform GUI workflow manager which aids scientists in their computational research. It can serve as a local application on you computer or as a centralized server for collaborative teamwork. Integrated with a 2D infinite canvas, Horus presents itself as a modular working environment. Its autonomous blocks enable customizable and distributable workflows via an accessible Python API.
It is a pipeline designed to be executed daily and gather data around disinformation. To this purpose, the pipeline takes the fact-checks that have been identified that day (through the APIs of Iberifier and Google), extracts keywords from these fact-checks, and uses these keywords to look for information related to these claims in digital media (MyNews service) and social networks (Twitter).
The tool is based on a Bayesian statistical modelling framework and provides functionalities for performing exploratory analysis, fitting spatiotemporal models, testing and validating the predictive ability of the models, and forecasting the probability of a disease outbreak. Additionally, the tool includes functions to visualise outputs for all modelling steps and final predictions. Its flexible design allows users to input both observed (long-lag) and forecast (short-lag) hydrometeorological indicators, such as drought and flood indicators, and output the probability of an outbreak of a given climate-sensitive disease (e.g., dengue, malaria, or leptospirosis) several months in advance.
Workflows for the analysis of immunological genomes.
This software analyzes a large volumes of posts to massively profile potential criminals on social networks and media.
IOAnalyzer analises the I/O flow of a local node, and detects I/O patterns. With this information IOAnalyzer changes to the best I/O scheduler improving performance.
BSD 3-Clause License
Iterative high-throughput calibration of epidemiological models is a workflow that performes optimization-via-simulation and uses time series on reported cases to find candidate values for unknown epidemiological parameters.
Library to validate JSON files
Solderpad Hardware License
Lagarto Hun is a segmented scalar processor with in-order execution using the RISC-V open-source instruction set.
In collaboration between BSC and CIC-IPN, originally Lagarto Hun was a RV64IMA core with a 5-stage pipeline in order, supporting privileged ISA v1.11. As a first improvement, we have added support for single and double precision floating point operations, becoming Lagarto Hun into a RV64GCV + (subset RVV0.7.1) core, commonly named as RV64G for general purpose core.
It may be used for embedded systems and education.
Solderpad Hardware Licence (Version 2.1)
One of the main goals in processor design is achieving high performance. Various proposals from academia and industry have explored innovative ideas leveraging Moore's Law, balancing complexity and performance. SUperscalar processors execute multiple instructions per clock cycle, effectively using instruction-level parallelism, dynamic scheduling, efficient memory management, and specialized accelerators.
libPRISM is designed to tune several hardware knobs of a processor. It is an interposition library for Linux platforms to manage hardware knobs at execution time without user intervention. There is no need to recompile software to use it. Easily extendable for different programming models and manage different hardware knobs. In its current form, libPRISM optimize the SMT level and the data prefetcher hardware knobs on a POWER8 processor for OpenMP applications.
LightDock is a protein-protein, protein-peptide and protein-DNA docking framework written in Python.
This technology integrates a list of benchmarks for LLMs and computes their performance. Includes medical benchmarks, general purpose as well as bias and toxicity ones.
Apache License (Version 2.0) with LLVM exceptions
This is a version of the LLVM compiler that fulfills the EPI requirements when it comes to vectorization for RISC-V Vector Extension. This compiler includes an implementation of the EPI-specific intrinsics and the vectorizer for the RISC-V Vector Extension. This compiler enables vectorization of applications using the RISC-V Vector Extension either using the mentioned intrinsics or the automatic vectorization mechanism. The vectorization mechanism can be triggered automatically or explicitly requested via pragma annotations such as #pragma clang loop or #pragma omp simd.
Apache License (Version 2.0) with LLVM exceptions
Version of the LLVM compiler that implements the OmpSs.