Tux: Efficient Drop-in Networking for Database Systems
Xinjing Zhou, Viktor Leis, Xiangyao Yu, Michael Stonebraker
摘要
Modern high-performance OLTP systems are increasingly bottle-necked by kernel network stack and the high cost of dispatching requests to database worker threads. While kernel-bypass TCP stacks improve performance, they often sacrifice compatibility and robustness, in addition to leaving performance opportunities on the table due to TCP's byte-stream interface. We present Tux , a kernel-bypass networking stack designed for database systems that achieves high performance without giving up compatibility and robustness. TUX addresses these challenges by (1) introducing a message-based transport protocol that decouples reliability from in-order delivery and natively preserves message boundaries without framing or copy overhead inherent in byte-stream interface; (2) providing a flexible pushdown abstraction that lets database engines execute DBMS-specific logic closer to the NIC to avoid context-switch overhead and to exploit message-based interface; and (3) leveraging eBPF/XDP to reuse well-maintained kernel NIC drivers for compatibility and operational ease. We implement TUX in a library called LIBTUX, offering both zero-change "compatibility" mode and minimal-change "pushdown" mode. Our implementation, LIBTUX, evaluated on VoltDB, Redis, ScyllaDB, Memcached, and LeanStore, improves throughput up to 2.3x, reduces median and 99th percentile latencies by up to 2.6x and 4.7x, compared to existing kernel-bypass systems with minimal modifications to the evaluated DBMSes.
问问这篇 Paper
智能体会读完全文。
Lune 把这篇 Paper 索引到了每一个公式,引用它的顶会 Paper 也一样。你提问,回答直接引用原文。
它引用的顶会 Paper13
- Caladan: Mitigating Interference at Microsecond TimescalesJoshua Fried, Zhenyuan Ruan, Amy Ousterhout, Adam BelayOSDI 2020 · 被引用 213 次
- The Demikernel Datapath OS Architecture for Microsecond-scale Datacenter SystemsIrene Zhang, Amanda Raybuck, Pratyush Patel, Kirk Olynyk 等SOSP 2021 · 被引用 83 次
- BMC: Accelerating Memcached using Safe In-kernel Caching and Pre-stack ProcessingYoann Ghigoff, Julien Sopena, Kahina Lazri, Antoine Blin 等NSDI 2021 · 被引用 79 次
- Making Kernel Bypass Practical for the Cloud with JunctionJoshua Fried, Gohar Irfan Chaudhry, Enrique Saurez, Esha Choukse 等NSDI 2024 · 被引用 57 次
- CoroBase: Coroutine-Oriented Main-Memory Database EngineYongjun He, Jiacheng Lu, Tianzheng WangVLDB 2021 · 被引用 42 次
相关 Paper
- DINT: Fast In-Kernel Distributed Transactions with eBPFYang Zhou, Xingyu Xiang, Matthew Kiley, Sowmya Dharanipragada 等NSDI 2024 · 被引用 40 次
- Tigger: A Database Proxy That Bounces With User-BypassMatthew Butrovich, Karthik Ramanathan, John Rollinson, Wan Shen Lim 等VLDB 2023 · 被引用 17 次
- Low-Latency Communication for Fast DBMS Using RDMA and Shared MemoryPhilipp Fent, Alexander van Renen, Andreas Kipf, Viktor Leis 等ICDE 2020 · 被引用 44 次
- In-Network Support for Transaction TriagingTheo Jepsen, Alberto Lerner, Fernando Pedone, Robert Soulé 等VLDB 2021 · 被引用 19 次
- The Art of Latency Hiding in Modern Database EnginesKaisong Huang, Tianzheng Wang, Qingqing Zhou, Qingzhong MengVLDB 2024 · 被引用 23 次
