Building consistent transactions with inconsistent replication
Authors:
Open Access
- 4 October 2015
- conference paper
- Published by Association for Computing Machinery (ACM)
- Vol. 35 (4) , 263-278
Abstract
Application programmers increasingly prefer distributed storage systems with strong consistency and distributed transactions (e.g., Google's Spanner) for their strong guarantees and ease of use. Unfortunately, existing transactional storage systems are expensive to use -- in part because they require costly replication protocols, like Paxos, for fault tolerance. In this paper, we present a new approach that makes transactional storage systems more affordable: we eliminate consistency from the replication protocol while still providing distributed transactions with strong consistency to applications. We present TAPIR -- the Transactional Application Protocol for Inconsistent Replication -- the first transaction protocol to use a novel replication protocol, called inconsistent replication, that provides fault tolerance without consistency. By enforcing strong consistency only in the transaction protocol, TAPIR can commit transactions in a single round-trip and order distributed transactions without centralized coordination. We demonstrate the use of TAPIR in a transactional key-value store, TAPIR-KV. Compared to conventional systems, TAPIR-KV provides better latency and throughput.Keywords
Funding Information
- NSF (CNS-0963754, CNS-1217597, CNS-1318396, CNS-1420703,CNS-1518702)
This publication has 29 references indexed in Scilit:
- CassandraACM SIGOPS Operating Systems Review, 2010
- BigtableACM Transactions on Computer Systems, 2008
- Lower bounds for asynchronous consensusDistributed Computing, 2006
- Consensus on transaction commitACM Transactions on Database Systems, 2006
- Optimistic replicationACM Computing Surveys, 2005
- Providing Persistent Objects in Distributed SystemsLecture Notes in Computer Science, 1999
- The temporal logic of actionsACM Transactions on Programming Languages and Systems, 1994
- Providing high availability using lazy replicationACM Transactions on Computer Systems, 1992
- On optimistic methods for concurrency controlACM Transactions on Database Systems, 1981
- A Majority consensus approach to concurrency control for multiple copy databasesACM Transactions on Database Systems, 1979