Outputs

The research outputs from PAStor will be disseminated through top scientific venues in the HPC and storage fields and released as an open-source prototype. The prototype will provide the first building block towards a novel storage architecture suited for exascale computing infrastructures. The project will include both senior researchers and students who are working on PhD and MSc thesis focused on the scope of the project.

This page will be updated during the project duration.


Publications

  • Miranda M, Esteves T, Portela B, Paulo J. S2Dedup: SGX-enabled Secure Deduplication. ACM International Systems and Storage Conference (SYSTOR). 2021

  • Macedo R, Tanimura Y, Haga J, Chidambaram V, Pereira J., Paulo J. PAIO: A Software-Defined Storage Data Plane Framework. ArXiv – Computing Research Repository (CoRR), 2021.


Talks

  • PAStor’s presentation at UT Austin Portugal Program’s workshop “Better prepared: Strengthening societies through international scientific collaboration”. Collocated with Encontro Ciência’21 in Lisbon, Portugal.

Prototypes

  • S2Dedup leverages intel SGX to enable privacy-preserving deduplication at untrusted servers. The open-source prototype, built with SPDK, supports multiple secure schemes. Deduplication is more and more used to optimize the performance and reduce the storage costs of Big Data applications (e.g., Deep Learning frameworks) that deal with large quantities of information. Our solution enables this technique at third-party infrastructures, such as HPC centers, without compromising the privacy of users’ sensitive data.