Demonstration venue · read-only. Every page can be browsed; the buttons that would change it are switched off. Create an account to run TaxoReview on your own data. A simulation from public publication records: the people listed were not invited and did not agree to review.
? Tutorial for this page

Ertza Warraich

dblp:299/8878 · DBLP profile ↗
← Back
3ranked-venue papers
1as first author
3since 2021 · last 2026
0000-0002-6670-5076ORCID · corroborated

Domains — the database's venue-derived domains; a paper can count in several

Computer networks · 2 · 1 first-author · 2 since 2021Security and privacy · 1 · 1 since 2021

Expertise — from the expertise taxonomy: the topics of the expert's papers under the CCF categories. A weight counts papers with recency: 1 for a paper about the topic, 0.3 when the topic is its context, halved every five years.

Computer architecture, parallel and distributed computing, and storage systems
1 paper
High-performance computing · 40% Cloud and datacenter computing · 40% Distributed systems · 20%
Network and information security
1 paper
Systems and software security · 100%
Software engineering, system software, and programming languages
1 paper
Program analysis · 100%

Topics — the 8 heaviest of 9, each with the papers that count most for it

TopicWeightPapersLastEvidence papers
High-performance computing › collective communication
all-reduce
0.912025
OptiReduce: Resilient and Tail-Optimal AllReduce for Distributed Deep Learning in the Cloud · NSDI 2025
Cloud and datacenter computing › resource management
cloud resource management
0.912025
OptiReduce: Resilient and Tail-Optimal AllReduce for Distributed Deep Learning in the Cloud · NSDI 2025
High-performance computing
collective communication
0.912025
OptiReduce: Resilient and Tail-Optimal AllReduce for Distributed Deep Learning in the Cloud · NSDI 2025
Distributed systems › distributed machine learning
distributed deep learning
0.912025
OptiReduce: Resilient and Tail-Optimal AllReduce for Distributed Deep Learning in the Cloud · NSDI 2025
Cloud and datacenter computing › quality of service
tail latency
0.912025
OptiReduce: Resilient and Tail-Optimal AllReduce for Distributed Deep Learning in the Cloud · NSDI 2025
Systems and software security
vulnerability discovery
0.612022
FuzzUSB: Hybrid Stateful Fuzzing of USB Gadget Stacks · SP 2022
Program analysis
static analysis
0.212022
FuzzUSB: Hybrid Stateful Fuzzing of USB Gadget Stacks · SP 2022
Program analysis
symbolic execution
0.212022
FuzzUSB: Hybrid Stateful Fuzzing of USB Gadget Stacks · SP 2022

Methods — techniques the papers use, named apart from their topics

symbolic execution · 1.1static analysis · 1.1state machine extraction · 1.1fuzzing · 1.1
YearPublicationVenuePosition
2026 Towards Network-Efficient Cross-Regional Inference via Learned Activation Compression
abstract
Large transformer models are increasingly deployed across geographically distributed GPU clusters due to capacity, cost, and locality constraints. When inference is partitioned across sites, intermediate activations must be transmitted over wide area network (WAN) links at each partition boundary, introducing significant communication overhead. We present Feather, a system that reduces this overhead by compressing intermediate activations before transmission and reconstructing them before downstream layers resume execution.
Regan McDonald, Marilyn Rego, Ertza Warraich, Annus Zulfiqar, Muhammad Shahbaz 0001
SIGCOMM3
2025 OptiReduce: Resilient and Tail-Optimal AllReduce for Distributed Deep Learning in the Cloud
Ertza Warraich, Omer Shabtai, Khalid Manaa, Shay Vargaftik, Yonatan Piasetzky, Matty Kadosh, Lalith Suresh 0001, Muhammad Shahbaz 0001
NSDI1
2022 FuzzUSB: Hybrid Stateful Fuzzing of USB Gadget Stacks
abstract
Universal Serial Bus (USB) is the de facto protocol supported by peripherals and mobile devices, such as USB thumb drives and smart phones. For many devices, USB Type-C ports are the primary interface for charging, file transfer, audio, video, etc. Accordingly, attackers have exploited different vulnerabilities within USB stacks, compromising host machines via BadUSB attacks or jailbreaking iPhones from USB connections. While there exist fuzzing frameworks dedicated to USB vulnerability discovery, all of them focus on USB host stacks and ignore USB gadget stacks, which enable all the features within modern peripherals and smart devices. In this paper, we propose FuzzUSB, the first fuzzing framework for the USB gadget stack within commodity OS kernels, leveraging static analysis, symbolic execution, and stateful fuzzing. FuzzUSB combines static analysis and symbolic execution to extract internal state machines from USB gadget drivers, and uses them to achieve state-guided fuzzing through multi-channel inputs. We have implemented FuzzUSB upon the syzkaller kernel fuzzer and applied it to the most recent mainline Linux, Android, and FreeBSD kernels. As a result, we have found 34 previously unknown bugs within the Linux and Android kernels, and opened 8 CVEs. Furthermore, compared to the baseline, FuzzUSB has also demonstrated different improvements, including $ 3\times$ higher code coverage, $ 50\times$ improved bug-finding efficiency for Linux USB gadget stacks, $ 2\times$ higher code coverage for FreeBSD USB gadget stacks, and reproducing known bugs that could not be detected by the baseline fuzzers. We believe FuzzUSB provides developers a powerful tool to thwart USB-related vulnerabilities within modern devices and complete the current USB fuzzing scope.
Kyungtae Kim, Taegyu Kim, Ertza Warraich, Byoungyoung Lee, Kevin R. B. Butler, Antonio Bianchi, Jing (Dave) Tian
SP3