Andrea Arpaci-Dusseau
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Andrea Arpaci-Dusseau has a total of 39 co-authors. Here are the top co-authors by number of shared publications:
- Remzi H. Arpaci-Dusseau
University of Wisconsin–Madison, Madison, Wisconsin, USA
13 shared publications - Jing Liu
Microsoft Research, Beijing, China
4 shared publications - Suli Yang
Google
3 shared publications - Aditya Jain
Department of Computer Sciences, University of Wisconsin-Madison, 1205 University Avenue, Madison, WI 53706, USA
2 shared publications - Kai Mast
Provable Inc., 5470 Kietzke Ln Ste 300, Reno, NV 89511, USA
2 shared publications - Shawn (Wanxiang) Zhong
University of Wisconsin-Madison, Madison, Wisconsin, USA
2 shared publications
Latest Publications
- Open Access
- 28 September 2026
- Published by Association for Computing Machinery (ACM)
- p. 882-897
TensorHub: Scalable and Elastic Weight Transfer for LLM RL Training
- 28 September 2026
- Published by Association for Computing Machinery (ACM)
- p. 882-897
Open AccessTensorHub: Scalable and Elastic Weight Transfer for LLM RL Training
- Open Access
- 28 September 2026
- Published by Association for Computing Machinery (ACM)
- p. 353-378
Don't Let AI Agents YOLO Your Files: Information and Control in Agent-Native Filesystems
- 28 September 2026
- Published by Association for Computing Machinery (ACM)
- p. 353-378
Open AccessDon't Let AI Agents YOLO Your Files: Information and Control in Agent-Native Filesystems
- Open Access
Data-Centric Serverless Computing with LambdaStore
LambdaStore is a data-centric serverless platform that breaks the split between stateless functions and external storage in classic cloud computing platforms. By scheduling serverless invocations near data instead of pulling data to compute, LambdaStore substantially reduces the state access cost that dominates today’s serverless workloads. Leveraging its transactional storage engine, LambdaStore delivers serializable guarantees and exactly-once semantics across chains of lambda invocations—a capability missing in current Function-as-a-Service offerings. We make three key contributions: (1) an object-oriented programming model that ties function invocations with its data; (2) a transaction layer with adaptive lock granularity and an optimistic concurrency control protocol designed for serverless workloads to keep contention low while preserving serializability; and (3) an elastic storage system that preserves the elasticity of the serverless paradigm while lambda functions run close to their data. Under read-heavy workloads, LambdaStore lifts throughput by orders of magnitude over existing serverless platforms while holding end-to-end latency below 20 ms.
Open AccessData-Centric Serverless Computing with LambdaStore
LambdaStore is a data-centric serverless platform that breaks the split between stateless functions and external storage in classic cloud computing platforms. By scheduling serverless invocations near data instead of pulling data to compute, LambdaStore substantially reduces the state access cost that dominates today’s serverless workloads. Leveraging its transactional storage engine, LambdaStore delivers serializable guarantees and exactly-once semantics across chains of lambda invocations—a capability missing in current Function-as-a-Service offerings. We make three key contributions: (1) an object-oriented programming model that ties function invocations with its data; (2) a transaction layer with adaptive lock granularity and an optimistic concurrency control protocol designed for serverless workloads to keep contention low while preserving serializability; and (3) an elastic storage system that preserves the elasticity of the serverless paradigm while lambda functions run close to their data. Under read-heavy workloads, LambdaStore lifts throughput by orders of magnitude over existing serverless platforms while holding end-to-end latency below 20 ms.
- Open Access
- 11 December 2025
- Published by MDPI AG in Preprints.org
Data-Centric Serverless Computing with LAMBDASTORE
LAMBDASTORE is a new serverless platform with an integrated storage engine tailored for stateful serverless workloads. Its compute-storage co-design colocates serverless functions with their associated data, yielding significant performance gains. It also leverages the transaction interface of its storage engine to provide serializable workflows and exactly-once semantics. This paper presents the design of LAMBDASTORE and introduces three key contributions. First, it adopts an object-oriented model in which functions are bundled with their associated data, enabling function execution to be scheduled directly at the data’s location. Second, the storage layer provides efficient transaction processing by dynamically adjusting lock granularity and employing a customized optimistic concurrency control protocol. Third, to enable colocation without sacrificing elasticity, the system supports data migration and lightweight replication at the granularity of individual objects. Experiments show that LAMBDASTOREoutperformsconventional serverless platforms, especially in read-heavy workloads. In such settings, LAMBDASTORE achieves throughput orders of magnitude higher than existing systems, while maintaining average end-to-end latencies below 20 ms.
- 11 December 2025
- Published by MDPI AG in Preprints.org
Open AccessData-Centric Serverless Computing with LAMBDASTORE
LAMBDASTORE is a new serverless platform with an integrated storage engine tailored for stateful serverless workloads. Its compute-storage co-design colocates serverless functions with their associated data, yielding significant performance gains. It also leverages the transaction interface of its storage engine to provide serializable workflows and exactly-once semantics. This paper presents the design of LAMBDASTORE and introduces three key contributions. First, it adopts an object-oriented model in which functions are bundled with their associated data, enabling function execution to be scheduled directly at the data’s location. Second, the storage layer provides efficient transaction processing by dynamically adjusting lock granularity and employing a customized optimistic concurrency control protocol. Third, to enable colocation without sacrificing elasticity, the system supports data migration and lightweight replication at the granularity of individual objects. Experiments show that LAMBDASTOREoutperformsconventional serverless platforms, especially in read-heavy workloads. In such settings, LAMBDASTORE achieves throughput orders of magnitude higher than existing systems, while maintaining average end-to-end latencies below 20 ms.
- Open Access
- 13 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 63-72
Tidying Up the Address Space
- 13 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 63-72
Open AccessTidying Up the Address Space
- Open Access
- 8 September 2025
- Published by Association for Computing Machinery (ACM)
- p. 202-215
PANGOLIN: a Comprehensive Testing Framework for Configuration-Rich Key-Value Stores
- 8 September 2025
- Published by Association for Computing Machinery (ACM)
- p. 202-215
Open AccessPANGOLIN: a Comprehensive Testing Framework for Configuration-Rich Key-Value Stores
- Open Access
- 30 March 2025
- Published by Association for Computing Machinery (ACM)
- p. 21-41
Revealing the Unstable Foundations of eBPF-Based Kernel Extensions
- 30 March 2025
- Published by Association for Computing Machinery (ACM)
- p. 21-41
Open AccessRevealing the Unstable Foundations of eBPF-Based Kernel Extensions
- Open Access
Foreactor: Exploiting Storage I/O Parallelism with Explicit Speculation
We introduce explicit speculation, a variant of I/O speculation technique where I/O system calls can be parallelized under the guidance of explicit application code knowledge. We propose a formal abstraction -- the foreaction graph -- which describes the exact pattern of I/O system calls in an application function as well as any necessary computation associated to produce their argument values. I/O system calls can be issued ahead of time if the graph says it is safe and beneficial to do so. With explicit speculation, serial applications can exploit storage I/O parallelism without involving expensive prediction or checkpointing mechanisms. Based on explicit speculation, we implement Foreactor, a library framework that allows application developers to concretize foreaction graphs and enable concurrent I/O with little or no modification to application source code. Experimental results show that Foreactor is able to improve the performance of both synthetic benchmarks and real applications by significant amounts (29%-50%).
Open AccessForeactor: Exploiting Storage I/O Parallelism with Explicit Speculation
We introduce explicit speculation, a variant of I/O speculation technique where I/O system calls can be parallelized under the guidance of explicit application code knowledge. We propose a formal abstraction -- the foreaction graph -- which describes the exact pattern of I/O system calls in an application function as well as any necessary computation associated to produce their argument values. I/O system calls can be issued ahead of time if the graph says it is safe and beneficial to do so. With explicit speculation, serial applications can exploit storage I/O parallelism without involving expensive prediction or checkpointing mechanisms. Based on explicit speculation, we implement Foreactor, a library framework that allows application developers to concretize foreaction graphs and enable concurrent I/O with little or no modification to application source code. Experimental results show that Foreactor is able to improve the performance of both synthetic benchmarks and real applications by significant amounts (29%-50%).
- Open Access
- 8 July 2024
- Published by Association for Computing Machinery (ACM)
- p. 15-22
Shadow Filesystems: Recovering from Filesystem Runtime Errors via Robust Alternative Execution
We present Robust Alternative Execution (RAE), an approach to transparently mask runtime errors in performance-oriented filesystems via temporarily executing an alternative shadow filesystem. A shadow filesystem has the primary goal of robustness, achieved through a simple implementation without performance optimizations and concurrency while adhering to the same API and on-disk formats as the base filesystem it enhances. While the base performance-oriented filesystem may contain bugs, the shadow implementation is formally verified, leveraging advancements in the verification of low-level systems code. In the common case, the base filesystem executes and delivers high performance to applications; however, when a bug is triggered, the slow-but-correct shadow takes over, updates state correctly, and then resumes the base, thus providing high availability.
- 8 July 2024
- Published by Association for Computing Machinery (ACM)
- p. 15-22
Open AccessShadow Filesystems: Recovering from Filesystem Runtime Errors via Robust Alternative Execution
We present Robust Alternative Execution (RAE), an approach to transparently mask runtime errors in performance-oriented filesystems via temporarily executing an alternative shadow filesystem. A shadow filesystem has the primary goal of robustness, achieved through a simple implementation without performance optimizations and concurrency while adhering to the same API and on-disk formats as the base filesystem it enhances. While the base performance-oriented filesystem may contain bugs, the shadow implementation is formally verified, leveraging advancements in the verification of low-level systems code. In the common case, the base filesystem executes and delivers high performance to applications; however, when a bug is triggered, the slow-but-correct shadow takes over, updates state correctly, and then resumes the base, thus providing high availability.
- Open Access
- 6 March 2023
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Storage
- Vol. 19 (2) , 1-47
Principled Schedulability Analysis for Distributed Storage Systems Using Thread Architecture Models
In this article, we present an approach to systematically examine the schedulability of distributed storage systems, identify their scheduling problems, and enable effective scheduling in these systems. We use Thread Architecture Models (TAMs) to describe the behavior and interactions of different threads in a system, and show both how to construct TAMs for existing systems and utilize TAMs to identify critical scheduling problems. We specify three schedulability conditions that a schedulable TAM should satisfy: completeness, local enforceability, and independence; meeting these conditions enables a system to easily support different scheduling policies. We identify five common problems that prevent a system from satisfying the schedulability conditions, and show that these problems arise in existing systems such as HBase, Cassandra, MongoDB, and Riak, making it difficult or impossible to realize various scheduling disciplines. We demonstrate how to address these schedulability problems using both direct and indirect solutions, with different trade-offs. To show how to apply our approach to enable scheduling in realistic systems, we develop Tamed-HBase and Muzzled-HBase, sets of modifications to HBase that can realize the desired scheduling disciplines, including fairness and priority scheduling, even when presented with challenging workloads.
- 6 March 2023
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Storage
- Vol. 19 (2) , 1-47
Open AccessPrincipled Schedulability Analysis for Distributed Storage Systems Using Thread Architecture Models
In this article, we present an approach to systematically examine the schedulability of distributed storage systems, identify their scheduling problems, and enable effective scheduling in these systems. We use Thread Architecture Models (TAMs) to describe the behavior and interactions of different threads in a system, and show both how to construct TAMs for existing systems and utilize TAMs to identify critical scheduling problems. We specify three schedulability conditions that a schedulable TAM should satisfy: completeness, local enforceability, and independence; meeting these conditions enables a system to easily support different scheduling policies. We identify five common problems that prevent a system from satisfying the schedulability conditions, and show that these problems arise in existing systems such as HBase, Cassandra, MongoDB, and Riak, making it difficult or impossible to realize various scheduling disciplines. We demonstrate how to address these schedulability problems using both direct and indirect solutions, with different trade-offs. To show how to apply our approach to enable scheduling in realistic systems, we develop Tamed-HBase and Muzzled-HBase, sets of modifications to HBase that can realize the desired scheduling disciplines, including fairness and priority scheduling, even when presented with challenging workloads.