Global Stabilization for Causally Consistent Partial Replication

Global Stabilization for Causally Consistent Partial Replication
复制标题

因果一致部分复制的全局稳定性

DOI:
10.1145/3369740.3369795
复制
发表时间:
2020
期刊:
21st International Conference and Distributed Computing and Networking
影响因子:
--
通讯作者:
Vaidya, Nitin H.
Vaidya, Nitin H.
中科院分区:
--
文献类型:
--
作者:
Xiang, Zhuolun;Vaidya, Nitin H.

文献摘要

参考文献

被引文献

相似文献

因果一致的分布式存储系统由于具有提供高吞吐量的潜力而受到极大的关注。全局稳定化是一种用于在分布式多版本键值存储系统中实现因果一致性的技术,被GentleRain [6]和Cure [1]等以前的作品所采用。然而,具有全局稳定性的先前设计假设完全复制,其中数据集被分成分区,每个分区在所有数据中心处复制,并且每个客户端被限制为仅访问一个数据中心内的服务器。在本文中,我们提出了一个理论框架的全球稳定,以支持一般的部分复制与因果一致性,其中每个服务器可以存储任意子集的数据,每个客户端被允许与任何子集的服务器和迁移之间没有额外的延迟。我们提出了一种算法,实现因果一致性的分布式多版本的键值存储与一般的部分复制,我们的算法是最佳的远程更新的可见性延迟,即如何快速更新从远程服务器是可见的客户端,一般的部分复制。我们的算法的性能相比,以前的工作的仿真结果也提供。
Causally consistent distributed storage systems have received significant attention due to the potential for providing high throughput. Global stabilization is a technique established for achieving causal consistency in distributed multi-version key-value store systems, adopted by previous works such as GentleRain [6] and Cure [1]. However, previous designs with global stabilization assume full replication, where the set of the data is split into partitions with each partition replicated at all data centers, and each client is restricted to access servers within only one data center. In this paper, we propose a theoretical framework of global stabilization to support general partial replication with causal consistency, where each server can store an arbitrary subset of the data, and each client is allowed to communicate with any subset of the servers and migrate among them without extra delays. We propose an algorithm that implements causal consistency for distributed multi-version key-value stores with general partial replication, and our algorithm is optimal in terms of the remote update visibility latency, i.e. how fast update from a remote server is visible to the client, under general partial replication. Simulation results on the performance of our algorithm compared to the previous work are also provided.
DOI: --
发表时间: 2017-03
期刊: --
影响因子: --
作者:
Syed Akbar Mehdi;Cody Littley;Natacha Crooks;L. Alvisi;N. Bronson;Wyatt Lloyd
通讯作者: Syed Akbar Mehdi;Cody Littley;Natacha Crooks;L. Alvisi;N. Bronson;Wyatt Lloyd
利用写入语义来实现部分复制的因果对象
DOI: --
发表时间: 1998
期刊: International Euromicro Conference on Parallel, Distributed and Network-Based Processing
影响因子: --
作者:
M. Raynal;M. Ahamad
通讯作者: M. Ahamad
为地理分布式部分复制设计因果一致的协议
DOI: --
发表时间: 2015
期刊: PaPoC@EuroSys
影响因子: --
作者:
Tyler Crain;M. Shapiro
通讯作者: M. Shapiro
DOI: --
发表时间: 2017-02
期刊: --
影响因子: --
作者:
Chathuri Gunawardhana;Manuel Bravo;L. Rodrigues
通讯作者: Chathuri Gunawardhana;Manuel Bravo;L. Rodrigues
Wren:分区事务因果一致性数据存储中的非阻塞读取
DOI: --
发表时间: 2018
期刊: Dependable Systems and Networks
影响因子: --
作者:
Kristina Spirovska;Diego Didona;W. Zwaenepoel
通讯作者: W. Zwaenepoel