Harmonia: Near-Linear Scalability for Replicated Storage with In-Network Conflict Detection
Harmonia: Near-Linear Scalability for Replicated Storage with In-Network Conflict Detection
复制标题
DOI：
10.14778/3368289.3368301
复制
发表时间：
2019-04
期刊：
影响因子：
--
通讯作者：
Hang Zhu;Zhihao Bai;Jialin Li;Ellis Michael;Dan R. K. Ports;I. Stoica;Xin Jin
中科院分区：
文献类型：
--
作者：
Hang Zhu;Zhihao Bai;Jialin Li;Ellis Michael;Dan R. K. Ports;I. Stoica;Xin Jin
分布式存储采用复制来掩盖故障并提高可用性。然而，这些系统通常在一致性和性能之间表现出艰难的权衡。确保一致性会带来协调开销，因此系统吞吐量不会随着副本数量的增加而扩展。我们推出了 Harmonia，这是一种复制存储架构，它利用新一代可编程交换机的功能，通过在不牺牲一致性的情况下提供近线性可扩展性来消除这种权衡。为了实现这一目标，Harmonia 检测网络中的读写冲突，这使得任何副本都可以为没有挂起写入的对象提供读取服务。 Harmonia 以线路速率实现此功能，因此不会产生任何性能开销。我们在通过 Barefoot Tofino 交换机连接的商品服务器集群上实现了 Harmonia 原型，并将其与 Redis 集成。我们通过支持各种复制协议（包括主备份、链复制、Viewstamped Replication 和 NOPaxos）来展示我们方法的通用性。实验结果表明，对于 10 倍的复制因子，Harmonia 将这些协议的吞吐量提高了 10 倍，提供了接近线性的可扩展性，最高可达我们的测试台的极限。
Distributed storage employs replication to mask failures and improve availability. However, these systems typically exhibit a hard tradeoff between consistency and performance. Ensuring consistency introduces coordination overhead, and as a result the system throughput does not scale with the number of replicas. We present Harmonia, a replicated storage architecture that exploits the capability of new-generation programmable switches to obviate this tradeoff by providing near-linear scalability without sacrificing consistency. To achieve this goal, Harmonia detects read-write conflicts in the network, which enables any replica to serve reads for objects with no pending writes. Harmonia implements this functionality at line rate, thus imposing no performance overhead. We have implemented a prototype of Harmonia on a cluster of commodity servers connected by a Barefoot Tofino switch, and have integrated it with Redis. We demonstrate the generality of our approach by supporting a variety of replication protocols, including primary-backup, chain replication, Viewstamped Replication, and NOPaxos. Experimental results show that Harmonia improves the throughput of these protocols by up to 10 x for a replication factor of 10, providing near-linear scalability up to the limit of our testbed.