Sotabase

Career

· Assistant Professor, ETH Zürich2020–
· Research Scientist, Google Brain2019–2020
· Ph.D. Research Intern, Microsoft Research2015–2015
· Ph.D. Research Intern, Facebook2014–2014
· Masters in Electrical Engineering, Stanford University2013–2015
· Ph.D. in Electrical Engineering, Stanford University2013–2019
· Bachelors of Applied Science in Engineering Science, University of Toronto2009–2013

Publications (53)

IX: A Protected Dataplane Operating System for High Throughput and Low Latency
USENIX Symposium on Operating Systems Design and Implementation · 2014
471
cited
Pocket: Elastic Ephemeral Storage for Serverless Analytics
USENIX Symposium on Operating Systems Design and Implementation · 2018
362
cited
ReFlex: Remote Flash ≈ Local Flash
International Conference on Architectural Support for Programming Languages and Operating Systems · 2017
163
cited
SONIC: Application-aware Data Passing for Chained Serverless Applications
USENIX Annual Technical Conference · 2021
157
cited
Flash storage disaggregation
European Conference on Computer Systems · 2016
152
cited
Towards Demystifying Serverless Machine Learning Training
SIGMOD Conference · 2021
143
cited
tf.data: A Machine Learning Data Processing Framework
Proceedings of the VLDB Endowment · 2021
103
cited
Understanding Ephemeral Storage for Serverless Analytics
USENIX Annual Technical Conference · 2018
103
cited
Orion: Interference-aware, Fine-grained GPU Sharing for ML Applications
European Conference on Computer Systems · 2024
100
cited
Selecta: Heterogeneous Cloud Storage Configuration for Data Analytics
USENIX Annual Technical Conference · 2018
88
cited
OPTIMUSCLOUD: Heterogeneous Configuration Optimization for Distributed Databases in the Cloud
USENIX Annual Technical Conference · 2020
57
cited
Understanding Rack-Scale Disaggregated Storage
USENIX Workshop on Hot Topics in Storage and File Systems · 2017
49
cited
The IX Operating System
ACM Transactions on Computer Systems · 2016
47
cited
Corrigendum to “The IX Operating System: Combining Low Latency, High Throughput and Efficiency in a Protected Dataplane”
ACM Transactions on Computer Systems · 2017
40
cited
ML Training with Cloud GPU Shortages: Is Cross-Region the Answer?
EuroMLSys@EuroSys · 2024
39
cited
Plumber: Diagnosing and Removing Performance Bottlenecks in Machine Learning Data Pipelines
Conference on Machine Learning and Systems · 2021
38
cited
Cachew: Machine Learning Input Data Processing as a Service
USENIX Annual Technical Conference · 2022
30
cited
tf.data service: A Case for Disaggregating ML Input Data Processing
ACM Symposium on Cloud Computing · 2022
25
cited
RAIL: Predictable, Low Tail Latency for NVMe Flash
ACM Transactions on Storage · 2022
20
cited
Demystifying Cost-Efficiency in LLM Serving over Heterogeneous GPUs
International Conference on Machine Learning · 2025
19
cited
Sotabase