BFT Protocols Under Fire
BFT Protocols Under Fire
复制标题
DOI：
--
复制
发表时间：
2008-04
期刊：
影响因子：
5.6
通讯作者：
Atul Singh;Tathagata Das;Petros Maniatis;P. Druschel;Timothy Roscoe
中科院分区：
文献类型：
--
作者：
Atul Singh;Tathagata Das;Petros Maniatis;P. Druschel;Timothy Roscoe
最近关于拜占庭状态机复制的工作主要集中于在良性条件(局域网、同构副本、有限的崩溃故障)下具有更高性能的协议，而在典型的实际条件(广域网延迟、分组丢失、瞬时断开、共享资源)下的评估相对较少。这使得系统设计人员很难为真正的目标部署选择合适的协议。此外，大多数协议实现在运行时环境、加密库和传输的选择上有所不同，即使在类似的条件下也阻碍了直接的协议比较。我们提出了一个这样的协议的仿真环境，它结合了声明式网络系统和健壮的网络仿真器。协议可以从前者的高级说明性语言中的伪代码快速实现，而网络条件和通信包和密码原语的(测量的)成本可以插入到后者中。我们表明，所产生的模拟器忠实地预测了本地协议实现的性能，无论是公布的性能还是在我们的本地网络中测量的性能。我们使用模拟器来比较相同条件下的典型协议，并快速探索密码操作成本、工作负载、网络条件和故障变化的影响。例如，我们表明Zyzzyva在大多数但不是所有条件下都优于PBFT和Q/U等协议，这表明在实践中设计一刀切的协议可能很难，如果不是不可能的话。
Much recent work on Byzantine state machine replication focuses on protocols with improved performance under benign conditions (LANs, homogeneous replicas, limited crash faults), with relatively little evaluation under typical, practical conditions (WAN delays, packet loss, transient disconnection, shared resources). This makes it difficult for system designers to choose the appropriate protocol for a real target deployment. Moreover, most protocol implementations differ in their choice of runtime environment, crypto library, and transport, hindering direct protocol comparisons even under similar conditions. We present a simulation environment for such protocols that combines a declarative networking system with a robust network simulator. Protocols can be rapidly implemented from pseudocode in the high-level declarative language of the former, while network conditions and (measured) costs of communication packages and crypto primitives can be plugged into the latter. We show that the resulting simulator faithfully predicts the performance of native protocol implementations, both as published and as measured in our local network. We use the simulator to compare representative protocols under identical conditions and rapidly explore the effects of changes in the costs of crypto operations, workloads, network conditions and faults. For example, we show that Zyzzyva outperforms protocols like PBFT and Q/U undermost but not all conditions, indicating that one-size-fits-all protocols may be hard if not impossible to design in practice.