Xenic: SmartNIC-Accelerated Distributed Transactions
Xenic: SmartNIC-Accelerated Distributed Transactions
复制标题
DOI：
10.1145/3477132.3483555
复制
发表时间：
2021-10
期刊：
影响因子：
--
通讯作者：
Henry N. Schuh;Weihao Liang;Ming G. Liu;J. Nelson;A. Krishnamurthy
中科院分区：
文献类型：
--
作者：
Henry N. Schuh;Weihao Liang;Ming G. Liu;J. Nelson;A. Krishnamurthy
高性能分布式事务需要对数据库内存和协议元数据进行高效的远程操作。这种工作负载的高通信成本要求硬件加速。最近的研究已将RDMA应用于此，利用网络控制器来操纵主机内存，而无需消耗目标服务器上的CPU周期。然而，基本的读/写RDMA原语需要在数据结构和协议设计中进行权衡，从而限制了它们的优点。SmartCore是快速分布式事务的灵活替代方案，可将可编程计算核心和板载内存添加到网络接口。应用测量的性能特征，我们设计Xenic，一个SmartNIC优化的事务处理系统。Xenic采用异步聚合执行模型，以最大限度地提高网络和核心效率。Xenic的协同设计的数据存储实现了低开销的远程对象访问。此外，Xenic使用灵活的点到点通信模式在SmartMonitor之间最大限度地减少事务提交延迟。我们将Xenic与先前基于RDMA和RPC的事务系统与TPC-C，Retwis和Smallbank基准进行比较。我们对三个基准测试的结果显示，吞吐量提高了2.42倍、2.07倍和2.21倍，延迟减少了59%、42%和22%，同时每台服务器节省了2.3、8.1和10.1个线程。
High-performance distributed transactions require efficient remote operations on database memory and protocol metadata. The high communication cost of this workload calls for hardware acceleration. Recent research has applied RDMA to this end, leveraging the network controller to manipulate host memory without consuming CPU cycles on the target server. However, the basic read/write RDMA primitives demand trade-offs in data structure and protocol design, limiting their benefits. SmartNICs are a flexible alternative for fast distributed transactions, adding programmable compute cores and on-board memory to the network interface. Applying measured performance characteristics, we design Xenic, a SmartNIC-optimized transaction processing system. Xenic applies an asynchronous, aggregated execution model to maximize network and core efficiency. Xenic's co-designed data store achieves low-overhead remote object accesses. Additionally, Xenic uses flexible, point-to-point communication patterns between SmartNICs to minimize transaction commit latency. We compare Xenic against prior RDMA- and RPC-based transaction systems with the TPC-C, Retwis, and Smallbank benchmarks. Our results for the three benchmarks show 2.42x, 2.07x, and 2.21x throughput improvement, 59%, 42%, and 22% latency reduction, while saving 2.3, 8.1, and 10.1 threads per server.