Portrait of Oana Balmau

Oana Balmau

Associate Academic Member
Assistant Professor, McGill University, School of Computer Science
University of Toronto
Contractor, Nutanix
Research Topics
Computational Sustainability
Computer Systems
Distributed Systems
High Performance Computing
Machine Learning Systems
Parallel Computing
Website
Google Scholar
LinkedIn

Biography

Oana Balmau is an Assistant Professor in the School of Computer Science at McGill University, where she leads the DISCS Lab. She also holds a status-only appointment in the Computer Science department at the University of Toronto and is a part of MLCommons, where she co-founded MLPerf Storage, an open-source benchmark for storage on ML workloads. Her research focuses on storage systems and data management, with an emphasis on ML, data science, and edge computing workloads.

She completed her PhD in Computer Science at the University of Sydney, advised by Prof. Willy Zwaenepoel. Before her PhD, Oana earned her Bachelors and Masters degrees in Computer Science from EPFL, Switzerland. Oana won the CORE John Makepeace Bennet Award 2021 for the best computer science dissertation in Australia and New Zealand, an Honorable Mention for the ACM SIGOPS Dennis M. Ritchie Doctoral Dissertation Award 2021, as well as Best Paper Awards in the USENIX Annual Technical Conference (USENIX ATC) 2019, and the ACM/IEEE Symposium on Edge Computing (SEC) 2024. Before joining McGill, Oana worked with Nutanix, ABB Research, and HP Vertica.

Current Students

Jiaxuan Chen
PhD - McGill University
Co-supervisor :
Xue (Steve) Liu
Rahma Nouaji
PhD - McGill University
Website
Shubham Vashisth
PhD - McGill University
Website

Publications

Wagg: Cost-aware Aggregation of Windowing Operators in Stream Processing
Pritish Mishra
Ruoyu Deng
Alexandre da Silva Veith
Oana Balmau
Eyal de Lara
2026-06-19
International Conference on Mobile Systems, Applications and Services (published)
doi.org
FOLD: Fuzzy Online Deduplication for Very Large Evolving Datasets via Approximate Nearest Neighbor Search
Nelson Bore
Pritish Mishra
Constantin Adam
Eyal de Lara
Oana Balmau
Fuzzy deduplication is key to constructing large language model training corpora. However, classic Locality-Sensitive Hashing pipelines scal… (see more)e poorly as corpora grow and are ill-suited to continuous ingestion. We present FOLD (Fuzzy Online Deduplication), an online fuzzy deduplication system that delivers high recall and throughput for evolving datasets. FOLD maintains an incrementally updated HNSW index over admitted documents, retrieving a small, high-quality candidate neighborhood for each incoming document instead of repeatedly rebuilding global buckets or rescanning the accumulated corpus. To our knowledge, FOLD is the first online fuzzy deduplication system to use HNSW. However, applying Jaccard similarity out of the box causes score crowding, making graph traversal unreliable within a small number of steps. FOLD addresses this with a bitmap representation that provides a more discriminative, Jaccard-aligned signal during HNSW search. Across four LLM-scale datasets (LM1B, C4, RealNews, and Common Crawl), FOLD stays fast and accurate as the corpus grows: at the largest evaluated scales, it maintains 93-97% recall and achieves up to 2.09x higher throughput than competing alternatives, whose best recall reaches only 76%.
2026-06-01
arXiv (preprint)
doi.org
arxiv.org
A Pragmatic Approach to Learned Indexing in RocksDB: Targeted Optimizations with Minimal System Modification
Shubham Vashisth
Olivier Michaud
Bettina Kemme
Oana Balmau
Learned indexes have emerged as a promising alternative to traditional index structures, offering higher throughput and lower memory usage b… (see more)y approximating the cumulative key distribution function with lightweight models. Despite these benefits, adoption in production systems remains limited, partly because learned indexes that support concurrency and persistence as effectively as, e.g., the B+-Tree, do not yet exist, while many research prototypes introduce substantial complexity. In this paper, we investigate whether off-the-shelf learned indexes can be integrated into a production database with minimal storage-engine redesign. Using RocksDB as a case study, we exploit its separation between in-memory Memtables and immutable on-disk files to deploy specialized indexes at each level. We show that directly applying existing learned indexes is insufficient under write-heavy workloads because frequent Memtable replacement prevents models from fully adapting. To address this, we introduce a reuse mechanism that preserves structural knowledge across Memtable instances. At the storage level, we replace RocksDB's disk index with a learned index without modifying the storage layer or read path. We further adapt a read-only learned index to be block-aware, enabling worst-case single-I/O lookups. We implement these techniques in MountDB, an extension of RocksDB. Experiments on large-scale workloads with diverse data distributions and access patterns show up to 1.5X higher write throughput and 2.1X higher read throughput than state-of-the-art systems, demonstrating that established learned indexes can be integrated into production systems with minimal overhead and substantial performance benefits.
2026-05-21
arXiv (preprint)
doi.org
arxiv.org
The Cost of Expertise: Understanding MoE Decode Performance
Sami Abuzakuk
Oana Balmau
Jiaxuan Chen
Anne-Marie Kermarrec
Rafael Pires
Ramya Prabhu
Martijn de Vos
2026-04-27
EuroMLSys @ European Conference on Computer Systems (published)
doi.org
MinatoLoader: Accelerating Machine Learning Training Through Efficient Data Preprocessing
Rahma Nouaji
Stella Bitchebe
Ricardo Macedo
Oana Balmau
Machine learning (ML) frameworks, such as PyTorch and TensorFlow, rely on data loaders to preprocess data before feeding it to accelerators.… (see more) When preprocessing is inefficiently pipelined, GPUs can remain idle over long periods of time, leading to substantial training delays. For example, PyTorch's default data loaders can cause up to 76% GPU idleness. A key bottleneck is the variability in preprocessing time across samples within the same dataset. Existing data loaders are oblivious to this variability, training all samples uniformly. In this case, a single slow sample can stall the entire batch, causing head-of-line blocking.
2026-04-25
European Conference on Computer Systems (published)
doi.org
arxiv.org
PostLearn: Towards A Learned Index For PostgreSQL
Abrar Fuad
Shubham Vashisth
Oana Balmau
Bettina Kemme
2026-04-25
Challenges and Opportunities of Efficient and Performant Storage Systems @ European Conference on Computer Systems (published)
doi.org