Uso de cookies

En las páginas web de la Universidad Carlos III de Madrid utilizamos cookies propias y de terceros para mejorar nuestros servicios mediante el análisis de sus hábitos de navegación. Al continuar con la navegación, entendemos que se acepta nuestra política de cookies. "Normas de uso"

Research Lines


ARCOS is a dynamic research group that is currently conducting research in various HPC areas.  Our research activities are oriented to the development of new software for large-scale parallel and distributed systems, These activities cover the development and optimization of parallel and distributed applications, real time, reliable designs, and high performance computing, including  cross-layer optimizations of HPC storage I/O stack, parallel file systems,  I/O acceleration of scientific workflows, auto-tuning of parallel I/O based on machine learning, dynamic monitoring of HPC infrastructures, convergence of HPC and Bigdata software stacks, and resource allocation elasticity in HPC and clouds.

All the former knowledge has been applied to several scientific and industrial areas, such as fluid mechanics, reconstruction of 3D medical images, Big Data Analytics, finance computing, and simulations involving complex systems, trying to combine real world engineering with cutting-edge research topics.  As a result, ARCOS has strong relations with a portfolio of major European laboratories and companies in areas such as Informatics, Bioengineering and Aerospace.

Moreover, ARCOS researchers are leading several European projects, such as COST Action IC1305 entitled “Network for Sustainable Ultrascale Computing” (NESUS,, with more than 70 institutions from 40 countries all around the world, or the projects REPARA ( and RePhrase ( that address the programmability for parallel systems with special attention to heterogeneous architectures.

Besides that, ARCOS is actively involved in the standardization of the C++ programming language (representing Spain in the ISO C++ standards committee).

Major ARCOS research lines:  

High-performance I/O and data-intensive computing.

Parallel programming models for application optimization.

  • Parallel Desing Patterns.
  • Design by Contract and High Integrity Software.
  • Programming Languages Design and Evolution.

Modeling and Simulation.