Research

Our research is currently divided in four main domains:

Efficient, Dependable, and Secure Storage

Storage systems are critical for the efficiency of data-centric applications (e.g., AI, analytics, databases) and large-scale services (e.g., Cloud, HPC). DSR designs new storage solutions and optimizations that meet the performance, scale, dependability, and security demands of traditional and emerging data-centric workloads.

Storage and OSs for Modern Data Centers

The emergence of novel hardware and the pressing need to decrease the carbon footprint of computing infrastructures are reshaping how data centers are built and managed. DSR is designing a new generation of storage and OS building blocks fitted for the performance, reliability, and energy consumption of modern infrastructures.

Benchmarking and Observability

Researching efficient and reliable storage solutions involves evaluating the performance, resilience, and security of these systems. DSR is committed to designing and developing advanced benchmarking and observability tools to accurately assess storage systems and data-centric applications, providing valuable insights for their improvement.

AI and Systems

Optimizing AI for systems and systems for AI workloads has become one of the most enticing routes for research as models and data grow and storage configurations may be suboptimal. DSR focuses on optimizing and improving systems for AI workloads while finding optimal configurations under volatile workloads.