In-Network Support for Transaction Triaging
Theo Jepsen, Alberto Lerner, Fernando Pedone, Robert Soulé, Philippe Cudré-Mauroux
Abstract
We introduce Transaction Triaging, a set of techniques that manipulate streams of transaction requests and responses while they travel to and from a database server. Compared to normal transaction streams, the triaged ones execute faster once they reach the database. The triaging algorithms do not interfere with the transaction execution nor require adherence to any particular concurrency control method, making them easy to port across database systems. Transaction Triaging leverages recent programmable networking hardware that can perform computations on in-flight data. We evaluate our techniques on an in-memory database system using an actual programmable hardware network switch. Our experimental results show that triaging brings enough performance gains to compensate for almost all networking overheads. In high-overhead network stacks such as UDP/IP, we see throughput improvements from 2.05× to 7.95×. In an RDMA stack, the gains range from 1.08× to 1.90× without introducing significant latency.
Ask about this paper
Your agent reads all of it.
Lune indexed this paper to the last equation, along with the top-tier papers that cite it. Ask a question and the answer quotes them.
Your agent calls
Luneget_paper_fulltext
Free to start. No credit card required.
Terminal
Install the CLIlune papers fulltext bab96422-6aa9-4e96-910c-683a27ed7a09Cited by top-tier papers4
- MTP: Transport for In-Network ComputingTao Ji, Rohan Vardekar, Balajee Vamanan, Brent E. Stephens et al.NSDI 2025 · 9 citations
- Pushing the Limits of In-Network Caching for Key-Value StoresGyuyeong KimNSDI 2025 · 9 citations
- GraphINC: Graph Pattern Mining at Network SpeedRana Hussein, Alberto Lerner, André Ryser, Lucas David Bürgi et al.SIGMOD 2023 · 7 citations
- High-Performance DBMSs with io_uring: When and How to Use ItMatthias Jasny, Muhammad El-Hindi, Tobias Ziegler, Viktor Leis et al.VLDB 2026 · 5 citations
Builds on4
- Programmable Calendar Queues for High-speed Packet SchedulingNaveen Kr. Sharma, Chenxingyu Zhao, Ming Liu, Pravein G. Kannan et al.NSDI 2020 · 119 citations
- Harmonia: Near-Linear Scalability for Replicated Storage with In-Network Conflict DetectionHang Zhu, Zhihao Bai, Jialin Li, Ellis Michael et al.VLDB 2020 · 58 citations
- Cheetah: Accelerating Database Queries with Switch PruningMuhammad Tirmazi, Ran Ben Basat, Jiaqi Gao, Minlan YuSIGMOD 2020 · 47 citations
- Scaling Distributed Machine Learning with In-Network AggregationAmedeo Sapio, Marco Canini, Chen-Yu Ho, Jacob Nelson et al.NSDI 2021
Related papers
- P4DB - The Case for In-Network OLTPMatthias Jasny, Lasse Thostrup, Tobias Ziegler, Carsten BinnigSIGMOD 2022 · 18 citations
- Xenic: SmartNIC-Accelerated Distributed TransactionsHenry N. Schuh, Weihao Liang, Ming Liu, Jacob Nelson et al.SOSP 2021 · 62 citations
- Chiller: Contention-centric Transaction Execution and Data Partitioning for Modern NetworksErfan Zamanian, Julian Shun, Carsten Binnig, Tim KraskaSIGMOD 2020 · 29 citations
- DINT: Fast In-Kernel Distributed Transactions with eBPFYang Zhou, Xingyu Xiang, Matthew Kiley, Sowmya Dharanipragada et al.NSDI 2024 · 40 citations
- Low-Latency Communication for Fast DBMS Using RDMA and Shared MemoryPhilipp Fent, Alexander van Renen, Andreas Kipf, Viktor Leis et al.ICDE 2020 · 44 citations
