Ciao! I’m Antonio 👋

I am a postdoctoral researcher at Data Intensive Applications & Systems Laboratory (DIAS) led by professor Anastasia Ailamaki.
I work on database systems and performance engineering at the intersection of query processing, compression, algorithms, and hardware acceleration.
My interests are:
- performance engineering for analytical systems
- hardware-accelerated analytics, including GPUs and on-chip accelerators
- storage in data management systems
- algorithms for indexing and compressing integers, strings, time series, and textual data

Previously I was a PhD student at the Department of Computer Science of the University of Pisa 🇮🇹, where I was member of the A³ lab led by professor Paolo Ferragina.
Selected Publications
-
[VLDB ’27]
Tile-Based Decompression in Compiled GPU Engines
Antonio Boffa, Ananya Gupta, Hamish Nicholson, Anastasia Ailamaki.
Proceedings of the VLDB Endowment, 20(2): 352–366, 2026.
Paper · PDF · Code · BibTeX -
[ADMS ’26] The Free Lunch Has a Queue: Characterizing On-chip Compression Accelerators for Analytics
Yi Jiang, Antonio Boffa, Hamish Nicholson, Anastasia Ailamaki.
17th International Workshop on Accelerating Analytics and Data Management Systems Using Modern Processor and Storage Architectures (ADMS @ VLDB), 2026.
Paper · PDF · Code · BibTeX -
[DaMoN ’25] The Effectiveness of Compression for GPU-Accelerated Queries on Out-of-Memory Datasets
Hamish Nicholson, Konstantinos Chasialis, Antonio Boffa, Anastasia Ailamaki.
21st International Workshop on Data Management on New Hardware (DaMoN @ SIGMOD), ACM, 2025.
DOI · PDF · BibTeX -
[JSS ’25] On the Compressibility of Large-Scale Source Code Datasets
Antonio Boffa, Roberto Di Cosmo, Paolo Ferragina, Andrea Guerra, Giovanni Manzini, Giorgio Vinciguerra, Stefano Zacchiroli.
Journal of Systems and Software, 227: 112429, 2025.
DOI · PDF · Code · BibTeX -
[ICDE ’25] Learned Compression of Nonlinear Time Series with Random Access
Andrea Guerra, Giorgio Vinciguerra, Antonio Boffa, Paolo Ferragina.
IEEE 41st International Conference on Data Engineering (ICDE), pp. 1579–1592, 2025.
DOI · PDF · Code · BibTeX -
[PhD Thesis ’24] Designing New Compressed Data Structures Using Data-Aware Approaches
Antonio Boffa.
PhD thesis, University of Pisa, 2024.
Link · BibTeX -
[IS ’24] CoCo-trie: Data-Aware Compression and Indexing of Strings
Antonio Boffa, Paolo Ferragina, Francesco Tosoni, Giorgio Vinciguerra.
Information Systems, 120: 102316, 2024.
DOI · PDF · Code · BibTeX -
[TALG ’22] A Learned Approach to Design Compressed Rank/Select Data Structures
Antonio Boffa, Paolo Ferragina, Giorgio Vinciguerra.
ACM Transactions on Algorithms, 18(3): 1–28, 2022.
DOI · PDF · Code · BibTeX
The full list of my publications is on Google Scholar.
Experiences


- Visiting PhD Student · DasLab, Harvard SEAS
Boston, Massachusetts, USA 🇺🇸 · Apr 2023 – Jun 2023
Conduct research activities focused on designing, implementing, and testing new data-aware approximate range filters against adversarial queries. The goal is to incorporate these range filters to real world LSM-tree based key-value stores like RocksDB.

- Software Developer Engineer Intern · Amazon AWS Redshift
Berlin, Germany 🇩🇪 · Jul 2021 – Oct 2021
Design, implement and test new theoretically grounded solutions to select the compression encodings for the columns of the tables inside a petabyte-scale cloud data warehouse (Redshift/Spectrum).

- Erasmus Exchange Student · University of Helsinki
Helsinki, Finland 🇫🇮 · Jan 2019 – Jun 2019
Services
Open-source contribution
-
ClickHouse: Identified, reproduced, and reported a bug in QBit vector search, used for semantic search over embeddings: on 1M DBpedia article embeddings, top-10 nearest-neighbour queries returned almost none of the true neighbours (near-zero recall) in the official release. Issue #89976, fixed upstream in PR #90485.
-
SereneDB: Built an experimental compressed full-text index (SDSL’s FM-index) for substring search. On 1M rows of ClickBench,
WHERE URL LIKE '%google%'went from 37.7 ms to 2.7 ms (14× faster) for 28% extra memory, close to a 3-gram index that needs 2.5× the disk space. Implementation, tests and reproducible benchmarks in PR #2224 (following #1980).
Reviewer:
- Proceedings of the VLDB Volume 20 (VLDB ‘27)
- IEEE International Conference on Data Engineering (ICDE ‘27)
- Symposium on Experimental Algorithms (SEA ‘26)
- ACM Transactions on Database Systems (TODS)
- Proceedings of the VLDB Volume 19 (VLDB ‘26)
- Journal of Supercomputing
- PLOS ONE ‘24
- Data Compression Conference (DCC ‘23)
- Symposium on Algorithm Engineering and Experiments (ALENEX ‘23)
Teaching & Supervision

Head Teaching Assistant, EPFL
- 01/02/2026 – 31/07/2026. Course: CS-300 Data-intensive systems
- 01/02/2025 – 31/07/2025. Course: CS-300 Data-intensive systems

Teaching Assistant, University of Pisa, Computer Science:
-
01/10/2023 – 30/01/2024 Course: Programming Lab II (2023/2024)
-
01/10/2022 – 30/01/2023 Course: Programming Lab II (2022/2023)
-
01/10/2021 – 30/01/2022 Course: Programming Lab II (2021/2022)
-
17/02/2020 – 15/07/2020 Course: Algorithms and laboratory (2019/2020)
Co-supervised theses:
- Martin Lettry. Cross-Engine Query Partitioning for Heterogeneous Hardware, Master Thesis @ IBM Zurich
- Arnaud Wei-Xin Shih. Systematic Performance Characterization of Query Engines on Heterogeneous Memory Systems. Master Thesis
- André Espírito Santo. Explore techniques to scale HNSW vector indexes with workload knowledge, Master Thesis @ Oracle Zurich
- Amey Kulkarni. GRASS: A Graph Serialization Framework for Querying and Visualizing Database Internal State, Master Thesis @ Oracle Zurich
Awards
Winner of the abroad study period scholarship 2023 from Fondazione ISSNAF (Italian Scientists and Scholars in North America Foundation)
Volunteering
Erasmus Student Network (ESN) is the biggest European association of university students, whose purpose is to promote and support international mobility exchanges between students. I’m a proud Honorary Member of ESN Pisa.