(Some people also call me nano.)
I completed my PhD in February 2014, under the supervision of Rachid Guerraoui and Viktor Kuncak. The title of my thesis is "Modularity in the Design of Robust Distributed Algorithms". It is about structuring distributed algorithms in independent modes that can be analyzed in isolation, making the understanding of such algorithms much easier. I formalized my work and proved theorems about it with the interactive proof assistant Isabelle/HOL. I also use TLA+ and its model-checker TLC to test my specifications and eliminate many bugs in them.
The TLA+ specifications appearing in my thesisThe Isabelle/HOL theories appearing in my thesis
A preliminary version of the work described in my thesis appeared in the paper: Guerraoui, Rachid, Viktor Kuncak, and Giuliano Losa. "Speculative linearizability", PLDI 2012. An updated version of the Isabelle/HOL theories can be found at the Archive of Forma Proofs, entry Abortable Linearizable Modules.
You can contact me by email at firstname.lastname@example.org
The fingerprint of my PGP key is D58B CCC5 28A5 F7CB 5B48 9CCF AE81 A995 EC2B 9EE4.
I am a research assistant professor in the Systems Software Research Group at Virginia Tech. My research is about improving the reliability of software systems through replication techniques, on multiple fronts: in the Popcorn Linux project, I am working on increasing the applicability of replication, by implementing it in the operating system to achieve transparency; in the Hyflow project, I am working on improving the performance of geo-replication algorithms; finally, in continuation of my PhD research, I am investigating how to effectively develop correct-by-construction distributed systems in the interactive theorem prover Isabelle/HOL.
Giuliano Losa, Sebastiano Peluso, Binoy Ravindran: Brief Announcement: A Family of Leaderless Generalized-Consensus Algorithms, PODC 2016. PDF and TLA+ specifications.
Sebastiano Peluso, Alexandru Turcu, Roberto Palmieri, Giuliano Losa, Binoy Ravindran: Making Fast Consensus Generally Faster, DSN 2016. PDF and TLA+ specifications.