Pedro F. Silvestre

Pedro F. Silvestre

PhD Student at Imperial College London

Large-Scale Data & Systems (LSDS) Group

Biography

I am a PhD student in the Large-Scale Data & Systems Group at Imperial College London, under the supervision of Dr. Peter Pietzuch. My research at Imperial will focus on the interplay between Dataflow Systems and Deep Learning. Before Imperial, I was a Research Engineer at the TU Delft’s Web Information Systems Group working on Clonos and a student at the NOVA School of Science and Technology.

I have also started the SysML@ICL seminar series and reading group.

Interests
  • Deep Learning (Systems)
  • Deep RL (Systems)
  • Dataflow Systems
  • Database Management Systems
  • Domain-specific Compilers
Education

  • PhD in ML Systems, 2021-Present

    Imperial College London

  • MSc in Computer Science, 2018-2020

    Nova School of Science and Technology

  • BSc in Computer Science, 2015-2018

    Nova School of Science and Technology

Recent Posts

Paper reviews, commentary and general tech stuff

On the Internals of Stream Processing Systems
A look at the internal architecture (design and thread model) of modern stream processing systems using Apache Flink as an example.
Pedro F. Silvestre
Sep 18, 2021 14 min read Technical Notes
On the Internals of Stream Processing Systems
Hello World!
An introduction to the website and what we will be doing. Paper reviews, tech ideas and such.
Pedro F. Silvestre
Mar 8, 2021 5 min read Meta
Hello World!

Recent Publications

Pedro F. Silvestre, Marios Fragkoulis, Diomidis Spinellis, Asterios Katsifodimos (2021). Clonos: Consistent Causal Recovery for Highly-Available Streaming Dataflows. Proceedings of the 2021 International Conference on Management of Data.

PDF Cite DOI

Popular Topics

"cloud computing" "consistency" "exactly-once" "high-availability" "stream processing" Academic Architecture Fault Tolerance Hugo Runtime Stream Processing Systems Systems

Send me an email!

This is the best way to reach me for professional matters