Andrea Lattuada (andrea.lattuada@inf.ethz.ch)
I'm a PhD Student in the Systems Group, Department of Computer Science at ETH Zürich. My advisor is Prof. Timothy Roscoe.
I'm supported by a Google PhD Fellowship.
My interests span various areas of systems and programming methodology. I am currently focusing on applications of the dataflow computational model, flow control mechanisms, distributed tracing, fault-tolerance in streaming systems, and language-supported enforcement of complex programming contracts.
Looking for a thesis topic?
We have several interesting projects and ideas suitable for Bachelor and Master thesis, or lab projects. If you are passionate about systems and want to work on state-of-the-art distributed systems, novel computational models, and fault-tolerance mechanisms, please send me an e-mail.
Contact info
CAB F 69, ETH Zürich, Universitätstrasse 6, Zürich, Switzerland
Office: +41 44 632 44 95
Publications
-
Conference Paper, 2020Shared Arrangements: practical inter-query sharing for streaming dataflows [Technical report], F. McSherry, A. Lattuada, M. Schwarzkopf, T. Roscoe[to appear]Proceedings of the VLDB Endowment
-
Conference Paper, August 2019Megaphone: Latency-conscious state migration for distributed streaming dataflows, M. Hoffmann, A. Lattuada, F. McSherry, V. Kalavri, J. Liagouris, T. RoscoeProceedings of the VLDB Endowment, Volume 12, 2018-2019 (VLDB 2019)
-
Workshop Paper, June 2018Latency-conscious dataflow reconfiguration, M. Hoffmann, F. McSherry, A. LattuadaProceedings of the 5th ACM SIGMOD Workshop on Algorithms and Systems for MapReduce and Beyond (BeyondMR'18)
-
Conference Paper, April 2018SnailTrail: Generalizing critical paths for online analysis of distributed dataflows [Technical report], M. Hoffmann, A. Lattuada, J. Liagouris, V. Kalavri, D. Dimitrova, S. Wicki, Z. Chothia, T. RoscoeProceedings of the 15th USENIX Symposium on Networked Systems Design and Implementation (NSDI'18)
-
Workshop Paper, June 2016[Workshop] Faucet: a user-level, modular technique for flow control in dataflow engines, A. Lattuada, F. McSherry, Z. ChotiaProceedings of the 3rd ACM SIGMOD Workshop on Algorithms and Systems for MapReduce and Beyond (BeyondMR'16)
-
October 2017[Poster] No-sync: coordination-free fault-tolerance in streaming, A. Lattuada, V. Kalavri, M. Hoffmann, F. McSherrySOSP'17 - Poster session
-
April 2017[Poster] Differential durability: fault-tolerant distributed differential computation, A. Lattuada, V. Kalavri, M. Hoffman, F. McSherryEurosys'17 - Poster session
-
April 2017[Poster] SnailTrail: Online Bottleneck Detection for your Dataflow, R. Sager, J. Liagouris, D. Dimitrova, A. Lattuada, V. Kalavri, Z. Chothia, M. Hoffmann, T. RoscoeEurosys'17 - Poster session
Selected talks
-
March 23rd, 2017Papers we love, Zürich chapter
-
April 7th, 2017Typeclasses — A type system constructScalar conf, Warsaw, Poland
-
September 30th, 2017RustFest, Zürich, Switzerland
-
February 3rd, 2019FOSDEM - Rust Devroom, Brussels, Belgium
Teaching assistant
-
Spring Semester 2017
-
Autumn Semester 2017, 2018
-
Spring Semester 2018
-
Autumn Semester 2019
The design and implementation of a lock-free ring-buffer with contiguous reservations
03 June 2019
This is the story of how James Munns and I designed and implemented (two versions!) of an high-perf lock-free ring-buffer for cross-thread communication. If any of those words look scary to you, don't fret, we'll explain everything from the basics.
Continue reading
A hammer you can only hold by the handle
05 November 2018
Today we're looking at the rust borrow checker from a different perspective. As you may know, the borrow checker is designed to safely handle memory allocation and ownership, preventing accessess to invalid memory and ensuring data-race freedom. This is a form of resource management: the borrow checker is tracking who's in charge of a chunk of memory, and who is currently allowed to read or write to it. In this post, we'll see how these facilities can be used to enforce higher-level API constraints in your libraries and software. Once you're familiar with these techniques, we'll cover how the same principles apply to advanced memory management and handling of other more abstract resources.
Continue reading