Distributed storage research at HASLab is motivated by the observation that storage systems are a cornerstone component for a wide range of research and industrial areas such as, cloud computing, high-performance computing, internet-of-things, databases, and artificial inteligence. Moreover, our research is driven by the need of such systems to provide efficient, scalable, resilient and secure solutions that can adapt to the exponential increase of digital information and to the different needs of very heterogeneous applications and infrastructures.
To this end, our research is focused on three main areas: Software-Defined Storage, Storage Benchmarking and Monitoring, Storage Optimization. In detail, with these three areas we intend to provide a new generation of programmable distributed storage systems that can adapt to heterogeneous storage workloads and deployment infrastructures, proper benchmarking tools for evaluating such systems, and novel storage functionalities and optimizations that can further improve the performance, reliability, and security of current solutions (see Research).
Our team is composed of researchers from the High-Assurance Software Laboratory (HASLab), mainly from the Distributed Systems field. HASLab is one of the integrated R&D centres of INESC TEC, a leading national associate laboratory, and a teaching and research subunit of the University of Minho, where it is headquartered.
Research at HASLab is anchored on a rigorous approach to three areas of Computer Science: Software Engineering, Distributed Systems and Cryptography and Information Security. The laboratory’s contributions to these areas range from fundamental research on formal methods and algorithms to applied research on tools and middleware that address real-world demands stemming from long-term collaborations with industry.
The paper Accelerating Deep Learning Training Through Transparent Storage Tiering was presented at the CCGrid'22 conference!24 February 2022
The paper PAIO: General, Portable I/O Optimizations With Minor Application Modifications was presented at the USENIX FAST'22 conference!1 January 2022
The PAStor and ACTPM projects were concluded with success!1 November 2021
The paper CaT: Content-aware Tracing and Analysis for Distributed Systems was accepted at the Middleware'21 conference!